Abstract
During newborn lung injury, excessive activity of lysyl oxidases (LOXs) disrupts extracellular matrix (ECM) formation. Previous studies indicate that TGFβ activation in the O2-injured mouse pup lung increases lysyl oxidase (LOX) expression. But how TGFβ regulates this, and whether the LOXs generate excess pulmonary aldehydes are unknown. First, we determined that O2-mediated lung injury increases LOX protein expression in TGFβ-stimulated pup lung interstitial fibroblasts. This regulation appeared to be direct; this is because TGFβ treatment also increased LOX protein expression in isolated pup lung fibroblasts. Then using a fibroblast cell line, we determined that TGFβ stimulates LOX expression at a transcriptional level via Smad2/3-dependent signaling. LOX is translated as a pro-protein that requires secretion and extracellular cleavage before assuming amine oxidase activity and, in some cells, reuptake with nuclear localization. We found that pro-LOX is processed in the newborn mouse pup lung. Also, O2-mediated injury was determined to increase pro-LOX secretion and nuclear LOX immunoreactivity particularly in areas populated with interstitial fibroblasts and exhibiting malformed ECM. Then, using molecular probes, we detected increased aldehyde levels in vivo in O2-injured pup lungs, which mapped to areas of increased pro-LOX secretion in lung sections. Increased activity of LOXs plays a critical role in the aldehyde generation; an inhibitor of LOXs prevented the elevation of aldehydes in the O2-injured pup lung. These results reveal new mechanisms of TGFβ and LOX in newborn lung disease and suggest that aldehyde-reactive probes might have utility in sensing the activation of LOXs in vivo during lung injury.
Keywords: aldehydes, bronchopulmonary dysplasia, lysyl oxidase, molecular probe, TGFβ
INTRODUCTION
Pulmonary development requires the precise coordination of biochemical, molecular, and cellular processes to enable the formation of functional gas-conducting and exchanging lung structures. The extracellular milieu plays a critical role in this process by enabling the intercellular communication that is necessary for the acquisition of a cell’s contextual awareness. This informs the regulation of cellular differentiation, proliferation, and migration, thereby contributing to orderly tissue development and organ formation. In addition, some of the molecules residing in the extracellular space provide a scaffolding that supports cellular and tissue development. The formation of these extracellular matrix (ECM) staging components into a supramolecular complex of fibers and networks requires the secretion, modification, assembly, and stabilization of tropoelastin monomers and collagen fibrils. For the developing lung, the assembly of these ECM components also imparts the acquisition of biophysical properties—elasticity and stiffness—that are required to support its physiologic function.
Three functional families of protein-modifying enzymes have a critical role in forming and stabilizing the ECM. They accomplish these activities, in part, by facilitating the crosslinking of nascent ECM proteins. First, secreted lysyl oxidases (LOXs) specialize in the catalytic deamination of the ε-amino group of peptidyl lysine and hydroxylysine substrates to form α-aminoadipic-δ-semialdehydes (for reviews, see Refs. 1 and 2). Lysyl oxidase (LOX) is the archetypal member of the LOXs family of enzymes, which also includes LOXL-1, LOXL-2, and LOXL-3. In the hydrophilic regions of secreted tropoelastin (3), and in specific telopeptide regions of secreted fibrillar collagen molecules, LOXs introduce reactive aldehyde moieties into lysine and hydroxylysine, forming allysine and hydroxyallysine through this process. Subsequently, the reactive aldehydes condense with others and with unmodified lysine residues to form intra- and intermolecular cross-links, and unique amino acids such as desmosine. These activities lead to the maturation of elastin and collagen molecules and enhance their structural and biophysical properties. But, studies suggest that secreted LOXs can also oxidize primary amines in plasma membrane proteins, thereby regulating signaling and cellular function (4). Moreover, in select cells, some forms of LOXs are rapidly taken up from the extracellular space and localize to the nucleus (4, 5). There, they might have a role in regulating gene expression (6–9). Second, lysyl hydroxylases also have a role in ECM formation. These enzymes form hydroxylysine, thereby providing the substrates used by LOXs to remodel ECM proteins. But, since the hydroxylysine residues can also serve as acceptors for sugars, lysyl hydroxylases also enable the glycosylation and the stabilization of some extracellular proteins (10, 11). Finally, transglutaminases catalyze the formation of isopeptide, Nϵ-(γ-l-glutamyl)-l-lysine bonds within and between polypeptide chains. These enzymes enhance the stability of several ECM proteins that play a role in ECM formation during pulmonary development (12, 13).
Transforming growth factor-β (TGFβ) is a key regulator of ECM formation and remodeling in the newborn lung. TGFβ has the ability to stimulate the expression and secretion of ECM precursor proteins, ECM cross-linking enzymes, and protease inhibitors (for review, see Ref. 14). Abnormal ECM formation appears to be a consistent feature of injury in the developing lung. Several studies have detected abnormal elastin and collagen structure in the injured lungs of mouse pups (15, 16) and preterm lambs (17–20), baboons (21), and infants (22–25). Accumulating evidence suggests that TGFβ has a central role in this dysregulation of ECM formation. For example, during hyperoxic (16, 26–29), hypoxic (30, 31), and ventilation-induced lung injury (20, 32–34) in the developing mouse and lamb lung, excessive TGFβ activation is associated with dysplastic ECM and alveolar development. Three lines of evidence support a primary role of TGFβ in causing this dysregulation of LOX expression. First, studies show that TGFβ is greatly activated in the O2-injured newborn lung (16, 26). Second, TGFβ has been determined to directly increase LOX expression in several cell types that model the newborn lung disease, including primary neonatal rat (35) and human lung fibroblasts (36) and mouse and human fibroblast cell lines (37). Third, studies demonstrate that antibody-mediated TGFβ neutralization blunts the increased in vivo expression of LOX, as well as other ECM-forming enzymes, in the O2-injured mouse pup lung and improves pulmonary ECM formation (16, 37, 38). This protective effect of TGFβ might be regulated, in part, by preventing excessive LOX expression and activation. This is because studies show that inhibition of LOXs during newborn lung injury decreases excessive elastin crosslinking, and improves ECM formation (39).
However, important details about how TGFβ and LOX dysregulate ECM formation during newborn lung injury remain unknown. For example, the intracellular mechanisms by which TGFβ increases LOX expression during newborn lung injury have not been determined. Moreover, whether the increase in LOXs during newborn lung injury causes accumulation of pulmonary aldehydes has not been reported. Also, whether increased pulmonary aldehyde levels might serve as a marker of excess activation of LOXs during newborn lung injury have not been investigated. Here, we report the results of experiments that provide mechanistic information about TGFβ and LOX in newborn lung disease.
MATERIALS AND METHODS
Antibodies and Reagents
In the immunoblotting work, pre-pro-, pro-, and mature LOX protein expression was detected using anti-LOX antibody (ab31238, Abcam, 1: 1,000), which reacts with the conserved COOH-terminus of all the enzyme forms. The anti-α-smooth muscle actin (αSMA) antibody [1A4 56856, 1:250 immunohistochemistry (IHC), 1:200 immunofluorescence (IF)] and anti-PDGF receptor alpha (PDGFRα) antibody (D1E1E 3174, 1:50) were obtained from Cell Signaling Technology (CST). TGFβ-Smad protein expression was detected using anti-Smad2 antibody (3103, CST, 1: 1,000), anti-Smad3 antibody (9523, CST, 1:1,000), and anti-Smad4 antibody (sc-7966, Santa Cruz, 1:200). Also, anti-glyceraldehyde 3-phosphate dehydrogenase (GAPDH; G8795, Sigma-Aldrich, 1: 5,000) antibody was used. Enzyme-conjugated secondary antibodies were obtained from the Jackson ImmunoResearch. In the immunofluorescence (IF) and IHC studies, the following antibodies were used: anti-LOX (ab31238, Abcam, 1:100 IF and 1:100 IHC), anti-Smad4 (sc-7966, Santa Cruz, 1:200 IF), anti-active TGFβ-1 antibody [LC (1-30); a gift from Kathleen Flanders and LalageWakefield, 7 µg/mL IHC], and a pan-TGFβ neutralizing antibody (1D11.16.8; BP0057, BioXCell). Isotype control antibodies were obtained from Abcam. Details about the specificity of all antibodies are provided in the Supplemental Table S1 (all Supplemental material is available at https://doi.org/10.17605/OSF.IO/YDB4V). Alexa Fluor-conjugated secondary antibodies were obtained from Thermo Fisher Scientific (TFS). Recombinant human TGFβ-1 (240-B, R&D Systems) was reconstituted in sterile 4 mM HCl containing 1 mg/mL bovine serum albumin. Actinomycin D (A9415) and β-aminopropionitrile fumarate salt (BAPN; A3134) were obtained from Sigma-Aldrich.
Mouse Pup Lung Injury Model
All experiments were approved by the Institutional Animal Care and Use Committee at the MGH. For the newborn lung injury model, pools of C57BL/6Ncr newborn pups of both sexes were randomly divided into litters of 7–8 pups and exposed to air or 85% O2 until postnatal day 21 (P21) with their dams. The sex of the pups was not taken into account. This is because studies using this O2-mediated lung injury model determined that sex does not influence pulmonary development (40) and pulmonary LOX expression at this age (Supplemental Fig. S1A). In addition, our pilot study showed that LOX mRNA expression was increased in 85% O2-exposed pup lungs without sex dimorphism (Supplemental Fig. S1B). Except for brief periods every other day when the dams were swapped between the gas exposure environments and the bedding was refreshed, the mice breathed the gas mixtures continuously. The synthetic O2-rich gas mixture was freshly blended using separately regulated flow meters (Cole-Palmer) and medical grade O2 and N2 gases and introduced into 19-L exposure chambers (073149932433, Sterilite). The fresh gas flow rate into the exposure chambers exceeded the calculated O2 consumption of the mice housed within it by >10-fold. The O2 treatment level was measured using an electrochemical analyzer (OM-25RME, Maxtec). For the inhibition of LOX enzyme activity, the dams drank water with or without BAPN ad lib commencing with the gas exposures. The BAPN dose was 3 mg/kg body wt based on the immediate postpartum dam weight and the estimated water consumption of 6 mL/day.
Cell Culture
Mouse embryo fibroblast cells (NIH3T3 fibroblasts) were purchased from and authenticated by the American Type Culture Collection (CRL-1658, ATCC). Primary mouse pup lung fibroblasts were isolated from the lung periphery of P10 C57BL/6Ncr mouse pups (Charles River) using type II collagenase (Worthington). The cells were cultured in complete medium, which consisted of DMEM containing 4.5 g/L glucose (hDMEM; 11995, Life Technologies), 10% (vol/vol) fetal bovine serum (SH300803, Hyclone), 0.29 mg/mL glutamine, 100 U/mL penicillin, and 100 µg/mL streptomycin. Serum starvation medium had the same basal medium and supplements but no serum. The cells were maintained in a 5% CO2-containing, 37°C humidified incubator. They were passaged using EDTA-trypsin before becoming confluent and the primary fibroblasts were used before the fourth passage. They were characterized by their protein expression of PDGFRα and αSMA, which are markers of interstitial myofibroblasts in the newborn lung (41). Cells transferred to chamber slides were fixed with 4% formaldehyde, permeabilized with 0.1% Triton X-100, and blocked with serum, each diluted in PBS, and then treated with anti-PDGFRα, Alexa Fluor 546-conjugated anti-rabbit IgG, anti-αSMA, and then Alexa Fluor 488-conjugated anti-mouse IgG antibodies, or nonimmune IgG before the secondary antibodies, before wide-field fluorescence imaging using an inverted microscope (TiE, Nikon), a cooled, high-sensitivity CCD camera system (Clara-E, Andor), a stable output light-emitting diode illumination source (Sola Light Engine, Lumencor), and control software (NIS-Elements, Nikon). Mycobacterium testing of our cells was not performed; no cellular toxicity or cytoplasmic DNA staining was detected in the cultures.
Immunohistochemistry
The LOX protein forms were detected and also co-localized with αSMA or active TGFβ-1 in mouse pup lung using specific antibodies, immunohistochemistry, and bright-field microscopy. After mouse pups were killed by intraperitoneal injection of 200 mg/kg body wt of pentobarbital, a hole was made in the diaphragm to permit the lungs to collapse, and a 0.6-mm outer diameter polyethylene tube (PE10, Harvard Apparatus) was secured in the trachea. Then the lungs were inflated with methacarn (methyl-Carnoy) fixative until the base reached the plane between the apex of the heart and the costophrenic angles and fixed in situ for 24 h. Subsequently, the lungs were dissected from the body, dehydrated, and embedded in paraffin. Six-micrometer thick sections of the left lung were obtained, cleared of paraffin, and rehydrated. When the localization of secreted LOX protein forms was desired, the lung sections were blocked with serum, and reacted with the LOX-reacting or control antibodies. To detect intracellular LOX, the sections were permeabilized using 0.1% Triton X-100 in PBS before blocking and reacting with the antibodies. Following washing unreacted antibodies with PBS, the sections were exposed to biotinylated secondary antibodies, enzyme-linked avidin-biotin complexes, colorimetric substrates, and counterstains. For the multiplex detection of LOX and αSMA or active TGFβ-1, the sections were permeabilized, and then treated with serum before being exposed to the anti-LOX or control antibodies. After extensive washing, the sections were exposed to biotinylated anti-rabbit antibody, enzyme-linked avidin-biotin complexes, and substrates. After the sections were washed, they were treated with an avidin biotin blocking reagent (SP2001, Vector Laboratories), reacted with either the anti-αSMA or anti-active TGFβ-1 antibodies LC (1-30) (42), exposed to biotinylated secondary antibodies, enzyme-linked avidin-biotin complexes, and substrates. Finally, the sections were counterstained with methyl green before being dehydrated and having a coverslip mounted. Subsequently, z-axis stack images were acquired using a microscope with a motorized stage (Ti-E, Nikon) and integrated CCD camera system (DS-Ri1, Nikon), and then extended focus images were constructed (43). In the microscopy studies reported in this work, images of the treatment- and control-study groups and of the control antibody-treatment and detection reactions were collected during the same experiment and under identical conditions.
Subcellular LOX and Smad4 Protein Expression Mapping and Quantification in Primary Mouse Lung Fibroblasts
The fibroblasts grown on 1.7-cm2 chamber slides were treated with 10 ng/mL TGFβ-1, or 10 µg/mL 1D11.16.8, an active TGFβ-neutralizing antibody as control, when 60% confluent. The cells were washed with PBS 16 h later, fixed with −20°C acetone, permeabilized with 0.1% Triton X-100 in PBS, blocked with 1.5% goat serum in PBS, and then incubated with the anti-Smad4 antibody overnight at 4°C. After reacting the cells with Alexa Fluor 488-conjugated secondary antibody they were incubated with the anti-LOX antibody for 1 h. After exposing the cells to Alexa Fluor 546-conjugated secondary antibody, coverslips were mounted (Vectashield with DAPI; H-1200, Vector Laboratories). The following method was employed to quantify nuclear LOX and Smad4 immunoreactivity in the fibroblasts. Random, nonoverlapping, and noncontiguous 224 µm by 168 µm wide-field fluorescent images that excluded the chamber edge but containing ∼70 cells in total were obtained for each treatment group using the fluorescence-imaging system described earlier. The nuclear region of interest was identified using images of the cells reacted with a DNA-reacting dye (DAPI), and the mean integrated intensity of the LOX and Smad4 immunoreactivity signal in that area was quantified using a custom script employing Python v 3.0 and Pillow v 6.0, an image processing library. During the cellular imaging, image processing, and data analysis, experimental bias was reduced by masking the involved investigator with respect to the cell treatment groups.
RNA Isolation and Quantification
RNA was extracted from cells using TRIzol reagent (15596026, Invitrogen) and dissolved in RNase-free H2O. The RNA quality was verified using spectroscopy (Nanodrop, TFS). cDNAs were synthesized using 1 µg RNA and PrimeScript RT reagents with a genomic DNA elimination reaction (RR047A, Takara). Quantitative (q)PCR was performed using ∼2% of the cDNA product, oligonucleotide primers for LOX (forward: 5′-CAG CCA CAT AGA TCG CAT GGT-3′, reverse: 5′-GCC GTA TCC AGG TCG GTT C-3′) (39), PAI-1 (forward: 5′-TCT GGG AAA GGG TTC ACT TTA CC-3′, reverse: 5′-GAC ACGCCA TAG GGA GAG AAG-3′) (44), and 18S rRNA (forward: 5′-GTA ACC CGT TGA ACC CCA TT-3′, reverse: 5′-CCA TCC AAT CGG TAG TAG CG-3′) (42), SYBR Premix EX TaqII (RR820, Takara), and a thermocycler instrument (QuantStudio 3, Applied Biosystems). The specificity of the PCR primers was validated empirically by examining the DNA melting profile. The relative mRNA expression levels were determined using the ddCT method. The relative mRNA expression levels were normalized to the mean level detected in the samples obtained from control reagent-treated cells. Each sample was processed in triplicate, but only the median value of these technical replicates was utilized in the analysis.
LOX Protein Expression Measurement
LOX protein expression was detected using immunoblotting. NIH3T3 fibroblasts grown on 9.5-cm2 wells were treated with or without 10 ng/µL TGFβ-1 in complete medium when 80% confluent. The cells were scraped into ice-cold lysis buffer containing 50 mM Tris·HCl (pH 7.4), 1 mM EDTA, 1 mM dithiothreitol, and protease and phosphatase inhibitors (78447, TFS) 24 h later. The lysates were then triturated using a small-bore needle and syringe, sonicated, and kept on ice. P21 C57BL mouse pups were euthanized, as described earlier, and the left lungs were dissected from the body, quick-frozen using liquid nitrogen and pulverized. The lung proteins were solubilized using the ice-cold lysis solution, described earlier. The protein concentrations were determined in the cell and lung lysates using bicinchoninic acid protein assay reagent (23227, TFS). Equal amounts of the proteins were resolved using SDS-PAGE and then electroblotted onto polyvinylidene difluoride membranes. The membranes were blocked using 5% nonfat dry milk in Tris-buffered saline (TBS) containing 0.05% Tween-20 and then exposed to primary antibodies. Afterward, immunocomplexes were detected using peroxidase-conjugated secondary antibodies and chemiluminescent substrates. Enhanced chemiluminescence signals were acquired using a cooled charge-coupled device (CCD) camera system (Chemi-Doc XRS, Bio-Rad).
Promoter-Reporter Plasmid Construction and Promoter Activity Measurement
pGL4-PmLOX, a plasmid containing the murine LOX promoter upstream of the firefly luciferase gene luc2 (Photinus pyralis), was constructed in the following manner. Double-stranded DNA corresponding to the murine LOX promoter (45), nucleotides 52,662,402–52,664,748 of chromosome 18 of the C57BL/6 mouse strain (GRCm39, seq ID NC000084.7), was produced using solid-phase DNA synthesis and then cloned into pMA-RQ (GeneArt). Subsequently, the plasmid was digested with XhoI, PciI, and PvuI and the promoter-containing fragment of the plasmid was ligated into pGL4.23 (E8411, Promega), which had been digested with XhoI and NcoI to remove the minimum promoter portion. The veracity of the plasmids was confirmed using restriction endonuclease mapping and Sanger DNA sequencing. For promoter activity measurement, NIH3T3 fibroblasts were seeded onto 2-cm2 wells at a density of 2.5 × 104 cells/cm2 and transfected with 1.25 µg pGL4-PmLOX and 0.5 µg pCMV-RL (E2261, Promega) when 60% confluent using Xfect transfection reagent (631318, Takara). The medium was refreshed 4 h later.
Canonical Smad RNAi
To knockdown the expression of canonical Smads, NIH3T3 fibroblasts were seeded onto 4-cm2 wells at a density of 1.1 × 104 cells/cm2 and transfected with 10 pmol esiRNAs when 60% confluent using LipoFectamine RNAiMAX Reagent (13778, Invitrogen). The endoribonuclease-prepared siRNA (esiRNA) targeting Smad2 (EMU022831), Smad3 (EMU014271), Smad4 (EMU016021), and enhanced green fluorescence protein (eGFP; EHUEGFP) were obtained from Sigma-Aldrich.
Allysine Assay
The allysine concentration in pup lung tissue from both sexes was quantified employing our previously described methods (46, 47) with the following modifications used to accommodate the relatively small size of the mouse pup lung. The left lung of the mouse pup was flash-frozen using liquid N2, pulverized, and suspended in 500 µL of PBS, pH 7.4. Immediately after 250 µL of lung homogenate, which represented ∼25 mg of the tissue, was transferred to a high-pressure reaction tube, 200 µL of water, 50 µL of 4 mM fluorescein, 500 µL of 12 M HCl, and 20 mg of sodium 2-naphthol-7-sulfonate were added, and the vials were capped. After hydrolyzing the tissue and labeling the proteins for 16–20 h at 110°C, the sample was cooled to room temperature and neutralized to pH 7.4. Then the allysine content was quantified using an analytical high-performance liquid chromatography (HPLC) with fluorescence detector (1260 Infinity II LC system, Agilent) using the following method. The sample was resolved using a 25 mm × 4 mm, 5 μm column (C8, Phenomenex) and mobile phases consisting of: solvent A, 0.1% trifluoroacetic acid (TFA) in water, and solvent B, 0.1% TFA in HPLC grade acetonitrile. The chromatography protocol consisted of 0–40 min; 5%–30% solvent B, 40–42 min; 30%–95% solvent B, 42–45 min; 95% solvent B, 45–47 min; 95.5% solvent B 47–50 min; 5% solvent B. In each case, solvent B was diluted with solvent A. The fluorescence signals in the resolved samples were determined using excitation (λex) and emission wavelengths (λem) 254 nm and 310 nm, respectively, between 0 and 40 min and 490 nm and 510 nm between 40 and 50 min elution times. The spectrophotometer peak areas were calibrated using the values obtained using the fluorescein standard eluted at 45 min. The allysine concentration was determined by comparing the measured values with those of a standard curve derived using a bisnaphthol derivative of allysine, which was synthesized using methods detailed previously (47), and corrected for lung mass.
Detection and Quantification of Extracellular Aldehydes in Lung Sections
For reaction with the biotinylated hydrazide probe, frozen lung sections from mouse pups of both sexes were treated with methacarn fixative, blocked for endogenous avidin and biotin, and exposed with and without 1 mM hydrazide-biotin probe (21339, TFS) in PBS for 2 hr. Subsequently, the sections were treated with Tris-buffered saline, to neutralize the probe, reacted with enzyme-linked avidin-biotin complexes, and then a colorimetric substrate (Vector Red) before being dehydrated and having a coverslip applied. For reaction with the dinitrophenyl-linked hydrazine probe, 6-µm thick methacarn-fixed lung sections from pups of both sexes were reacted with dinitrophenylhydrazine (DNPH; TCI D0846, VWR) in 2 N HCl for 30 min. After neutralizing the sections with TBS, they were blocked with serum and then exposed to anti-DNP antibody, biotinylated secondary antibody, enzyme-linked avidin-biotin complex, and a colorimetric substrate before being counterstained with methyl green, dehydrated, and having a coverslip applied. The DNPH signal was measured using a quantitative absorbance microscopy using method that was adapted from Ermert et al. (48). To determine the linear working range for the IHC reaction, the lung sections from O2-injured mouse pups were treated with DNPH, anti-DNP antibody, and the alkaline phosphatase-linked detection system and then reacted with Vector Red for up to 6 min. After the lung sections were processed as described earlier, four nonoverlapping images of the lung were obtained, avoiding large airways and blood vessels and the pleural surface, using Kohler illumination, a 525/20 nm bandpass filter (ET525/20m, Chroma) protected with heat glass (KG5, Chroma), a ×20 0.75 NA Plan Apo Ph2 lens, and the microscope and camera system described earlier. The images were segmented using an image analysis program, and the mean integrated intensities (mInt) were determined using a custom Python script. After ensuring that a linear signal was obtained during up to 6 min of Vector Red exposure, subsequent studies using the mouse pup lung sections were obtained using substrate exposure times within that period. The data were normalized to the average mInt detected in the air-treated, control pup lung sections.
Quantification of Gd-Labeled Aldehyde-Reactive Probe Levels in the Pup Lung
An aldehyde-reactive probe based on a gadoterate meglumine (Gd-DOTA) chelate conjugated to a hydrazide moiety, Gd-labelled aldehyde-reactive probe (Gd-Hyd), was synthesized as previously described (46). Mouse pups of both sexes that treated with air or 85% O2 as described earlier were injected with 100 mmol/kg Gd-Hyd in PBS intraperitoneally from 20 mM solutions determined by inductively coupled plasma mass spectrometry (ICP-MS, Agilent 8800-QQQ). The pups were killed 30 min later using sodium pentobarbital 200 mg/kg ip and the lungs and soleus and gastrocnemius muscles were obtained by dissection and flash-frozen using liquid N2. Subsequently, the tissue was pulverized, transferred to a reaction vessel, net weighed, and hydrolyzed overnight using 70% nitric acid. The Gd content in the samples was then measured using ICP-MS, as we previously described (49).
Data Analysis and Statistical Methods
Unless otherwise indicated, the experiments were repeated at least three times, and representative data from one experiment are shown. The data were analyzed using R (50). The data were first compared using a one-way model of ANOVA. When treatment variance was detected at a P ≤ 0.05 level using that method, a Wilcoxon test was used post hoc. When four or more comparisons are made, a Bonferroni correction of the P value was used, and they are indicated. Otherwise, P < 0.05 was considered to be significant.
RESULTS
TGFβ-Activated Interstitial Fibroblasts Exhibit Increased LOX Protein Expression in the O2-Injured Newborn Lung
The three TGFβ isoforms—TGFβ-1, TGFβ-2, and TGFβ-3—are products of different genes. But they signal often through the same receptor-mediated system. The TGFβ isoforms are secreted as inactive molecules and require extracellular activation before they can bind to their receptors on the surface of cells, stimulate intracellular signaling, and thereby regulate biological processes. For example, in cultured cells, studies indicate that activated TGFβ increases lung fibroblast transdifferentiation into myofibroblasts (51–53). TGFβ also stimulates LOX promoter activity and mRNA expression in cultured myofibroblast-like cells (54) as well as in rat and human lung fibroblasts (35, 36). We have previously demonstrated that TGFβ plays a direct role in increasing LOX expression in the O2-injured newborn lung (37). However, whether TGFβ stimulates LOX expression in interstitial fibroblasts in the injured newborn lung is unknown. Accordingly, we tested whether LOX protein expression maps to TGFβ-stimulated myofibroblasts in the O2-injured pup lungs. For this work, we employed multiplexed IHC and specific antibodies that detect LOX, activated TGFβ-1, and αSMA, a gene typically expressed by TGFβ-stimulated lung fibroblasts (51, 53, 55, 56).
During normal lung development, we detected αSMA protein expression primarily in muscle cells associated with the pulmonary vasculature and large airways. However, as shown in Fig. 1A, increased anti-αSMA antibody reactivity was detected in interstitial cells in the O2-injured mouse pup lung. This result is consistent with reports by others in which increased αSMA-expressing myofibroblasts were detected in the interstitium of the O2-injured mouse pup lung (41). Importantly, in our studies increased LOX protein expression was observed to co-localize with these interstitial αSMA-expressing myofibroblasts. This work suggests that myofibroblasts are the primary source of LOX expression in the parenchyma of the injured newborn lung. But we also tested whether activated TGFβ maps to the increased LOX protein expression (Fig. 1A). Although faint anti-active TGFβ-1 antibody reactivity was detected in the control pup lungs, it was greatly increased in the interstitium of lungs injured with 85% O2. These areas were similar to those harboring the LOX-expressing myofibroblasts discussed earlier. The increased LOX protein was observed to co-localize with the active TGFβ-1 in the injured lung interstitial cells. These studies support the notion that LOX gene expression is increased in TGFβ-activated fibroblasts in the injured newborn lung. The specificity of the antibody double-labeling methods is shown in Fig. 1B; no reagent cross-reactivity was detected when the indicated antibodies were omitted from the experimental protocol.
Figure 1.
Lysyl oxidase (LOX) protein expression is increased in transforming growth factor-β (TGFβ)-activated interstitial myofibroblasts in the O2-injured mouse pup lung. A: LOX and α-smooth muscle actin (αSMA) or active TGFβ-1 were mapped in peripheral lung sections of mouse pups treated with air or 85% O2 using specific antibodies, multiplexed immunohistochemistry (IHC), and colorimetric substrates (brown for LOX, red for αSMA and active TGFβ-1). The sections were counterstained with methyl green. O2-mediated injury increased the detection and co-localization of LOX with αSMA and with active TGFβ-1 proteins in interstitial fibroblasts. Typical images obtained are shown from the lungs of four pups in the air- and six pups in the O2-treated groups. B: the specificity of the multiplexed IHC system is supported by a lack of cross-reactivity in O2-injured pup lung sections exposed with (+) and without (−) the primary antibodies, as indicated.
Because we detected increased in vivo LOX protein expression in TGFβ-activated interstitial lung myofibroblasts, we next tested whether TGFβ has a direct role in modulating LOX gene expression in these cells. For this work, we tested whether a physiologic level of TGFβ stimulates LOX protein expression in isolated primary mouse pup lung fibroblasts. As shown in Fig. 2A, enzymatic dissociation of peripheral mouse pup lung tissue permits the isolation of cells that predominantly express PDGFRα and αSMA, in vivo markers of myofibroblasts in the developing lung (57). Others have determined that cultured fibroblasts secrete TGFβ (58), and that this TGFβ secretion likely increases αSMA protein expression in isolated pup lung fibroblasts. So the cultured cells in the control group were treated with a pan-specific, active TGFβ-neutralizing antibody. The TGFβ-stimulated fibroblasts were treated with 10 ng/mL TGFβ-1 because this level was detected in the tracheal effluents of newborns with O2-induced lung injury (59). This isoform of TGFβ was used because it transduces intracellular signals through the same receptor-mediated mechanisms as TGFβ-2 and TGFβ-3, and is readily available in a recombinant, active form. We confirmed that the TGFβ-1 stimulated intracellular canonical signaling by quantifying Smad4 nuclear localization. This is because when Smad4 binds to TGFβ-activated Smad2/3, it localizes with the complex to the nucleus (for review, see Ref. 60) and thereby serves as a hallmark of canonical TGFβ signaling. In this work, we determined that TGFβ directly increases LOX protein expression in the isolated mouse pup lung interstitial fibroblasts. As shown in Fig. 2B, TGFβ-1 treatment increased LOX immunoreactivity in addition to Smad4 nuclear localization in the pup lung fibroblasts. The apparent increase in protein expression in the TGFβ-treated fibroblast cells is supported by the densitometric measurement of nuclear epifluorescence intensity associated with the LOX- and Smad4-targeting antibody detection (Fig. 2C). The specificity of the antibodies is shown in Supplemental Fig. S2. It is interesting also to note that the LOX immunoreactivity was detected in the nucleus of the TGFβ-treated cells. Although the numbers of primary fibroblasts cultured from the mouse pup lungs were insufficient to permit determination of the nuclear LOX form, studies using cell lines suggest that the nuclear LOX represents the mature form of the enzyme that is cleaved and then taken up from the extracellular space (5, 61). Together, these data suggest that active TGFβ plays a direct role in regulating LOX expression in interstitial fibroblasts residing in the injured newborn lung.
Figure 2.
Transforming growth factor-β (TGFβ) directly increases lysyl oxidase (LOX) protein expression in interstitial fibroblasts isolated from the mouse pup lung. A: primary mouse pup lung interstitial fibroblasts were characterized using multilabeling indirect immunofluorescence and PDGF receptor alpha (PDGFRα)- and α-smooth muscle actin (αSMA)-reacting antibodies, or control antibodies, as indicated. Representative wide-field fluorescence microscopy images of three independent experiments are shown. B: TGFβ treatment directly increases LOX protein expression and Smad4 nuclear localization in primary mouse pup lung interstitial fibroblasts. The fibroblasts obtained from the peripheral mouse pup lung and cultured in complete medium were treated with either 10 µg/mL 1D11, a pan-TGFβ neutralizing antibody to inhibit autocrine TGFβ activity, and 0 ng/mL TGFβ-1 (Control), or with 10 ng/mL TGFβ-1 for 16 h. Subsequently, LOX and Smad4 protein expression were detected in permeabilized cells using antibodies and indirect immunofluorescence; the nuclei were identified using DNA-binding DAPI. Representative wide-field fluorescence microscopy cellular images of three experiments are shown. C: TGFβ increases quantitative indices of nuclear LOX and Smad4 compartmentation. The mean nuclear anti-LOX and Smad4 integrated intensities of control or TGFβ-1 treated mouse pup lung fibroblasts, as described earlier, were determined as detailed in materials and methods; ∼72 cells/group, *P < 0.05.
TGFβ Regulates LOX Gene Expression in Fibroblasts at a Transcriptional Level
Previous studies indicate that TGFβ increases LOX mRNA levels in mouse embryonic NIH3T3 fibroblast cells (37) and LOX protein levels in neonatal rat lung fibroblasts (35). But the regulatory mechanisms in fibroblasts are unknown. First, we determined that TGFβ rapidly regulates LOX mRNA levels in cultured fibroblasts. As shown in Fig. 3A, TGFβ-1 treatment increased LOX mRNA levels in the NIH3T3 fibroblasts within 4 h of treatment to a steady-state level that was sustained for at least 24 h. This increase in LOX mRNA was associated with an elevation in pre-pro-/pro-LOX protein expression in the fibroblasts. This is because an increase in immunoreactive protein with a molecular weight that is consistent with these LOX forms was detected in the lysates of the TGFβ-treated NIH3T3 fibroblasts (Fig. 3B). The absence of an immunoreactive protein with 32 and 23 kDa molecular weights consistent with cleaved mature LOX forms in the immunoblotting experiment suggests that pro-LOX is either not secreted by these cells, not cleaved after secretion, or not taken up from the cell surface or media after extracellular proteolysis. The lack of mature LOX in these TGFβ-treated NIH3T3 fibroblasts is supported also by our inability to detect nuclear LOX in them using the anti-LOX antibody and indirect immunofluorescence or a protein with the molecular weight of cleaved LOX using the anti-LOX antibody and immunoblotting (Fig. 3C). The specificity of the antibody is supported by its primary detection of proteins with molecular weights consistent with the LOX forms in full-length immunoblots (Supplemental Fig. S3), and work by others in which the forms were detected only following heterologous expression in cells that do not express LOX (see Supplemental Table S1).
Figure 3.
Transforming growth factor-β (TGFβ) increases lysyl oxidase (LOX) expression in mouse embryo fibroblasts. A: LOX mRNA expression is rapidly increased in TGFβ-treated mouse embryo NIH3T3 fibroblasts. Cells maintained in complete medium were treated with 0 (control) or 10 ng/mL TGFβ-1 commencing at 0 h and for the indicated time lengths. Subsequently, cell lysates were obtained and LOX mRNA and 18S rRNA levels were determined using specific primers and quantitative (q)PCR. The LOX mRNA % control was determined as described in materials and methods; it represents the TGFβ-regulated LOX mRNA expression level relative to the mean value detected in the control cells at the same time-point; n = 6/group; *P < 0.0125 vs. each of the other treatment durations. B: TGFβ increases the level of pre-pro-LOX/pro-LOX protein expression in NIH3T3 fibroblasts. The cells were treated without or with 10 ng/mL TGFβ-1 in complete medium for 24 h. Subsequently, they were lysed and the protein levels of LOX forms were determined using immunoblotting and an antibody that detects pre-pro-LOX/pro-LOX (42–57 kDa), bone morphogenetic protein 1(BMP1)-cleaved (32 kDa), and a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS)-cleaved (23 kDa) mature LOX. The glyceraldehyde 3-phosphate dehydrogenase (GAPDH) protein levels were also determined to assess sample loading and transfer; protein standards were used to determine the molecular weights. The blot image shown is representative of five independent studies. The pre-pro-LOX/pro-LOX protein levels, indexed to that of GAPDH, and relative mean level detected in the control samples were quantified using densitometry; n = 5/group; *P < 0.05. C: TGFβ increases cytosolic pre-pro-LOX/pro-LOX but not nuclear LOX detection in NIH3T3 fibroblasts. The fibroblasts were treated without or with TGFβ-1, as described earlier. LOX protein expression was detected in permeabilized cells using specific antibodies and indirect immunofluorescence; the nuclei were identified using DNA-binding DAPI. Representative wide-field fluorescence microscopy cellular images of three experiments are shown.
We next investigated the mechanisms by which TGFβ regulates LOX mRNA expression in fibroblasts. In addition, we studied the influence of serum on TGFβ regulation because previous studies suggest that serum regulates LOX gene transcription and mRNA stability in neonatal rat aortic smooth muscle cells (SMCs) (62). Similar to what was observed in the SMC, we determined that in NIH3T3 cell fibroblasts serum restriction increases LOX mRNA levels by regulating transcription and mRNA stability (Fig. 4A). This is because transcription inhibition by actinomycin (Act) D did not prevent the increase in LOX mRNA levels in the serum-restricted fibroblasts cells to the level that was detected in the cells treated with serum. But in both the serum-treated and -restricted fibroblast cells, TGFβ increased LOX mRNA levels. This appeared to be mediated primarily by an increase in LOX mRNA transcription because the LOX mRNA levels in the Act D-treated fibroblasts were not influenced by TGFβ exposure at either serum level.
Figure 4.
Transforming growth factor-β (TGFβ) regulates lysyl oxidase (LOX) expression at a transcriptional level in mouse embryo fibroblasts. A: NIH3T3 fibroblasts cultured in the presence (+) and absence (−) of 10% fetal bovine serum (FBS) were pretreated without or with 5 µg/mL actinomycin D (Act D) for 1 h, and then treated without or with 5 µg/mL Act D and 10 ng/mL TGFβ-1, as indicated, for 6 h. Subsequently, LOX mRNA levels were determined in cell lysates by quantitative (q)PCR; n = 6/group; *P < 0.05, **P < 0.0125. Top left: serum restriction increases LOX mRNA expression in fibroblasts. Top right: whereas transcription inhibition did not alter LOX mRNA levels in the FBS-treated cells, it prevented the increase in LOX mRNA levels caused by serum-restriction. The LOX mRNA levels in Act D-treated and serum-restricted cells are higher than that in the Act D-treated and FBS-treated ones, suggesting an increase in LOX mRNA stability during low-serum conditions. Bottom: in the FBS-treated fibroblasts, Act D treatment prevented the increase in LOX mRNA levels due to TGFβ-1 treatment. Moreover, TGFβ-1 treatment did not alter LOX mRNA levels when transcription was inhibited in serum-restricted fibroblasts. B: NIH3T3 fibroblasts were co-transfected with plasmids expressing Photinus pyralis luciferase driven by LOX promoter and plasmids expressing Renilla reniformis luciferase driven by a cytomegalovirus (CMV) promoter, as a control; 24 h later, the cells were treated without and with FBS and TGFβ-1 as described earlier and indicated. Cell lysates were obtained 24 h later and promoter activity was determined by measuring the luciferase activities. Luciferase levels were normalized to the average level detected in FBS-treated cells that were not treated with TGFβ-1. TGFβ was observed to increase LOX promoter activity in the fibroblasts independent of the serum treatment level; n = 5/group, **P < 0.0125.
Because the studies detailed earlier suggest that TGFβ and serum regulate mRNA levels in fibroblasts, we next tested whether they also stimulate LOX promoter activity. Although several putative cis-acting regulatory elements have been detected in the LOX promoter region for mouse, rat, and humans (45, 63, 64), their specific roles in activating or suppressing LOX gene expression has not been fully elucidated. Previously, TGFβ was observed to increase mouse LOX promoter activity in serum-treated fibroblasts (37). Accordingly, we tested how serum might influence TGFβ’s control of LOX promoter activity. We synthesized DNA consistent with the murine LOX promoter, including ∼2,350 bases of the 5′ untranslated region (UTR). This genomic area contains several possible control elements, TATA boxes, and cap sites (45). We determined that serum restriction increased LOX promoter activity; its activity was nearly twice that detected in the cells that were treated with serum (Fig. 4B). Moreover, TGFβ increased LOX promoter activity in both the serum-treated and -restricted fibroblasts; the magnitude of increase was the same independent of the serum level (percent increase LOX promoter activity during TGFβ treatment ± SD: serum treatment: 96 ± 14, serum restriction: 93 ± 45; n = 5 each group, P = 0.89). Together, these data indicate that TGFβ regulates LOX gene expression in fibroblasts at a transcriptional level independent of the serum-exposure conditions.
TGFβ Regulates LOX Gene Expression in Fibroblasts via Canonical Smad-Dependent Signaling
After secreted TGFβ is activated, it engages its type II receptor (TGFβ R2) and together they promote the recruitment, phosphorylation, and activation of the TGFβ type I receptors [TGFβ R1, activin-like kinase (ALK)4, and ALK5] in a heteromeric receptor complex (28, 52). The activated TGFβ R1 phosphorylates serine (S) residues in a COOH-terminal SSXS motif in Smad2 and Smad3 (Smad2/3). After recruiting Smad4, a co-Smad, the phosphorylated Smad2/3 migrate as a complex into the nucleus and regulate gene expression. Smad2/3 phosphorylation, the nuclear localization of Smad2/3 or Smad4, or the activation of Smad2/3-regulated genes serves as markers of canonical TGFβ signaling. However, TGFβ can also regulate gene expression via a variety of noncanonical pathways (65). In the newborn lung, for example, TGFβ has been demonstrated to stimulate several noncanonical signaling pathways (66, 67).
Although TGFβ increases LOX mRNA expression in several cell types, the intracellular regulatory pathways in fibroblasts are unknown. To determine how TGFβ regulates LOX gene expression in fibroblasts, we tested the requirement of Smad2/3 and Smad4 for TGFβ’s increase in LOX mRNA levels. For this work, we first examined whether RNA interference (RNAi) effectively decreases the expression of Smad2/3 and Smad4. We also tested the impact of Smad2/3 and Smad4 knockdown on their known mediation of the TGFβ-stimulated plasminogen activator inhibitor-1 (PAI-1) mRNA expression. As shown in Fig. 5A, transfection of NIH3T3 fibroblasts with Smad2-, Smad3-, and Smad4-targeting esiRNA decreases the protein expression of the genes in the cells. Moreover, the decrease in the Smad protein levels was associated with functional inhibition of their signaling activity. We know this because RNAi targeting of these Smads was associated with a blunting of TGFβ-stimulated increase in PAI-1 mRNA levels (Fig. 5B). Importantly, this functional knockdown of each of the TGFβ canonical Smads also blunted the increase in LOX mRNA levels caused by TGFβ treatment (Fig. 5C). In the control studies, the Smad-targeting RNAi itself did not reduce LOX mRNA levels. These studies detail the TGFβ regulatory mechanisms of LOX gene expression in fibroblasts and indicate, for the first time, that TGFβ regulates LOX mRNA expression via its canonical signaling pathway. Because our previous work determined that TGFβ plays a direct role in increasing LOX gene expression in the O2-injured pup lung (37), we next turned our attention to determining how newborn lung injury might regulate pulmonary pro-LOX secretion and activation, and its formation of extracellular aldehydes.
Figure 5.
Transforming growth factor-β (TGFβ) increases lysyl oxidase (LOX) mRNA expression through canonical Smad signaling pathway. A: RNAi decreases Smad2, Smad3, and Smad4 protein expression in NIH3T3 fibroblasts. Cells cultured in complete medium were transfected with esiRNA targeting Smad2, Smad3, Smad4, or enhanced green fluorescence protein (eGFP) as control (−), and 24 h later soluble cellular protein was obtained for assessment of relevant Smad protein expression by immunoblotting. Images representative of 3–4 experiments are shown. B: Smad2, Smad3, and Smad4 RNAi decrease canonical TGFβ-regulated plasminogen activator inhibitor-1 (PAI-1) mRNA expression in NIH3T3 fibroblasts. Cells were transfected with esiRNA targeting the indicated Smads or eGFP as indicated earlier. After 24 h, the cells were treated without or with 10 ng/mL TGFβ-1 for 4 h and cell lysates were obtained. Subsequently, the relative PAI-1 mRNA expression level was determined using quantitative (q)PCR; n = 6/group, *P < 0.0125. C: Smad2, Smad3, or Smad4 knockdown prevents TGFβ-mediated increase of LOX mRNA expression in NIH3T3 fibroblasts. Cells that were transfected with the indicated esiRNA and treated without or with TGFβ-1, as described earlier, were lysed and the relative LOX mRNA expression levels were determined using qPCR; n = 6/group, *P < 0.0125.
Pup Lung Injury Increases Pro-LOX Protein Secretion in Areas of Disrupted ECM and Alveolar Development
LOX appears to play an important role in ECM formation and newborn lung development. The expression of LOX is increased in the developing lung (37, 39) and targeted knockout of Lox decreases the levels and assembly of elastic fibers and inhibits alveolar formation in mouse pups (68, 69). LOX protein requires extensive post-translational modification to become activated and these data suggest that it is processed into its active form in the newborn lung. Accordingly, we tested whether pro-LOX is cleaved into its active, mature form in the developing lung. Using an antibody that detects the conserved COOH-terminal portion of the LOX forms, we detected immunoreactive proteins in solubilized mouse pup lung specimens with molecular weights that are consistent with the proteolytically cleaved, catalytically active forms of LOX, as well as the precursor, inactive pre-pro-/pro-LOX (Fig. 6A).
Figure 6.
Precursor lysyl oxidase (LOX) is processed to active forms in the newborn lung, and increasingly localizes to dysplastic alveolar structures in the injured lung. A: catalytically active LOX proteins are detected in mouse pup lungs. Solubilized proteins from P21 C57BL mouse pup lungs were resolved using PAGE, electroblotted onto charged membranes, and reacted with an antibody that detects the conserved COOH-terminal portion of LOX forms. Catalytically active LOXs formed by the proteases bone morphogenetic protein 1 (BMP1) and a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS) were detected in the lysates in addition to catalytically inactive pro-protein ones. The molecular weights (MWs) of the LOX forms are shown (italics) and mapped in comparison with protein molecular weights standards (right side of blot). LOX motifs: SP, signal peptide; Cu, copper-binding domain; LTQ, lysyl-tyrosyl quinone; n = 6. B: in the developing lung, scant levels of secreted pro-LOX protein/extracellular LOX are identified within the rim of secondary septa (*). But in the O2-injured lung, these extracellular LOX enzyme forms also appear more extensive within the tissue parenchyma (arrows). The secreted LOX forms were detected in nonpermeabilized, methacarn-fixed lung sections from mouse pups treated with air or 85% O2, as indicated, employing an anti-LOX antibody, immunohistochemistry, and a chromogenic substrate (brown). The sections were counterstained with hematoxylin. Typical images from 4 pups/group. C: in the O2-injured pup lungs, areas exhibiting an abnormally condensed and fimbriated elastin-staining pattern (double arrows) are detected, corresponding to the increased secreted pro-LOX/mature LOX shown in the previous panel. The lung elastin was identified using a resorcin fuchsin stain (purple); the lung sections were counterstained with tartrazine (yellow). Typical images of lung sections from 4 pups/group. D: in the O2-injured pup lung, secreted pro-LOX/extracellular LOX is associated with alveolar epithelial cells (arrowheads) in addition to interstitial cells in nonpermeabilized lung sections. Mature LOX is detected in the nuclei of these cells in permeabilized lung sections using the anti-LOX antibody and immunohistochemistry (IHC). The LOX forms were localized in nonpermeabilized and permeabilized O2-injured mouse pup lungs using the methods described earlier; methyl green counterstaining was used to permit nuclear mapping. Typical wide-field images obtained from the lungs of three air-treated and six O2-treated pups are shown.
LOX protein expression has been detected in the alveoli, bronchioles, and bronchi of newborn and adult mice (37, 70). Moreover, previous studies demonstrate that O2-mediated injury increases LOX protein levels in the newborn mouse lung (37, 39). However, it is unknown whether the secreted LOX forms localize to areas of ECM formation and alveolar development. Also, it is unknown whether newborn lung injury is associated with increased secretion of the LOX forms, particularly in parenchymal areas that exhibit abnormal ECM protein assembly and dysmorphic alveolar development. This is because the previous studies that mapped LOX immunoreactivity in the newborn lung employed permeabilized tissue, which does not distinguish between the secreted and intracellular forms of LOX.
To preferentially identify secreted LOX forms, we first identified a lung tissue fixation method that does not substantially permeabilize cellular plasma membranes. We reasoned that by using nonpermeabilized tissues, relatively thick lung sections, an antibody that reacts with an epitope conserved among all the LOX forms, and IHC we would be able to detect primarily the secreted pro-LOX and cleaved mature LOX forms residing in the extracellular space. For our testing how various fixatives might permeabilize sections of lung tissue, we employed an antibody that recognizes an abundant intracellular protein (glyceraldehyde 3-phosphate dehydrogenase, GAPDH). In this work, we observed that formaldehyde fixation was unsuitable for our experimental goals because it caused substantial cellular permeabilization. However, we determined that methacarn, a protein-precipitating tissue fixative, did not appreciably permeabilize the mouse pup lung specimens. Although the organic solvents in this fixative have the potential to extract phospholipids from membranes and increase cellular permeability, we observed only limited penetration of the antibodies reactive to GAPDH into the cells in the lung sections fixed with methacarn (Supplemental Fig. S4). Then using methacarn-fixed lung tissue and the anti-LOX antibody, we localized the secreted forms of the enzyme. For this work, we employed P21 mouse pup lungs because previous work showed that they exhibit high levels of LOX mRNA expression when injured with 85% O2 (37). As shown in Fig. 6B, scant amounts of secreted pro-LOX/extracellular mature LOX were identified in the normally developing lung, predominantly in the secondary septa of developing alveoli. This area was associated with elastin condensation and organization along the septa, as demonstrated by the use of an elastin stain (Fig. 6C). Elastin-reactive stain of the O2-injured lung demonstrated extension into the lung airspace, a pattern similar to that observed in the injured newborn lung of premature babies (23). In contrast with the normally developing lung, O2-injured pup lungs demonstrated an increased level of anti-LOX antibody immunoreactivity, which was localized to interstitial areas exhibiting dysmorphic ECM and alveolar septal formation. These areas of increased secreted LOX forms are similar to those where we previously detected the LOX protein expressing TGFβ-stimulated interstitial myofibroblasts (compare Fig. 6B and Fig. 1A). The specificity of the anti-LOX antibody is described earlier; the specificity of the secondary antibodies is shown in Supplemental Fig. S5.
We next tested whether the mouse pup lung injury is associated with the localization of LOX immunoreactivity to the nucleus of lung cells. We investigated this because studies using cultured cells show that when secreted pro-LOX is cleaved, the resulting mature and enzymatically active form of the enzyme can be taken up by cells from the extracellular space and transported to the nucleus (5, 70). In those studies, the nuclear form of LOX was shown to be solely the mature one. Previous studies have not detected nuclear LOX immunoreactivity in pulmonary cells of adult mice, although it has been detected in renal parenchymal cells and choroid plexus and ependymal cells in these animals (71). Because we detected increased extracellular LOX forms in the O2-treated mouse pup lung, we surveyed nuclear LOX compartmentation in injured lung sections. For this work, methyl green counter-staining was employed to enable better LOX protein detection within the nucleus. As shown in Fig. 6D, in nonpermeabilized injured mouse pup lung sections, secreted pro-LOX/mature LOX were detected around epithelial cells of the airways, in addition to interstitial cells in the distal lung parenchyma as shown earlier. Importantly, tissue permeabilization allowed cellular anti-LOX antibody penetration and the detection of immunoreactivity within the nuclei of these pulmonary cells. Although it is unknown whether the nuclear LOX immunoreactivity represents the mature form of the enzyme, these suggest that O2-mediated injury increases the secretion of pro-LOX and the nuclear compartmentation of LOX forms by cells in the newborn lung.
O2-Induced Injury Increases Aldehyde Levels in the Newborn Lung
Previous studies suggest that the enzyme activity of LOXs is increased in the O2-injured pup lung (37, 39). But it is unknown whether this produces levels of oxidative products that are in excess of the amount that can be consumed by protein cross-linking and thereby causes aldehyde accumulation in the lung. Previous studies do not provide insight in this issue. That is because they report that mouse pup lung aldehyde levels are both unchanged and increased following O2-mediated injury (72, 73). Increasingly, aldehydes are being recognized as a driver of disease processes (74). Accordingly, it is important to determine the role that LOX might play in aldehyde production in the injured newborn lung.
Because LOXs specialize in generating allysine from lysine by substituting an aldehyde for the ε-amino residue, we tested first whether allysine levels are increased in the O2-injured pup lung. To quantify the allysine levels, we optimized a highly selective assay that we previously used to assess their levels in fibrotic tissues (47) so that it could detect allysine levels in the smaller mouse pup lung tissue samples. This assay involves digesting lung proteins to liberate amino acids in the presence of a bisnaphthol derivative, which chemically modifies allysine with a fluorescent tag, resolving modified allysine from other amino acid species using HPLC, and quantifying their levels using spectroscopy. As shown in Fig. 7A, fluorescence spectroscopy of HPLC-resolved, acid-hydrolyzed pup lung specimens permitted the detection of bisnaphthol-modified allysine. Its identity was confirmed by comparing its retention time to that of a bisnaphthol-modified allysine standard. Importantly, we determined that O2-mediated injury caused an accumulation of allysine within the newborn lung (Fig. 7, A and B). This confirms that the activity of LOXs is increased in the injured newborn lung and that it generates allysine in levels that exceed their consumption by ECM protein cross-linking.
Figure 7.
O2-mediated injury increases allysine content in the mouse pup lung. Allysine levels were detected directly in the digests of lung tissues obtained from air- or 85% O2-treated mouse pups as described in materials and methods. The samples were digested in acid and reacted with 2-naphthol-7-sulfonate and fluorescein. Subsequently, the samples were neutralized, resolved using high-performance liquid chromatography (HPLC), and labeled compounds were detected using a fluorescence detector using arbitrary units (AU). Allysine levels were determined by comparing the signals detected in the samples with the ones obtained using allysine standards. A: typical fluorescence-detected HPLC traces of an air-exposed mouse pup lung digest without (black line) and subsequently with (gray line) allysine standard sample spiking (top); zoomed fluorescence-detected HPLC traces showing the difference in allysine peaks in lung lysates obtained from air- and 85% O2-exposed mouse pups (bottom). B: allysine levels are significantly increased in pup lungs injured with 85% O2; n = 7/group, *P < 0.05.
An In Vivo, Aldehyde-Reactive Molecular Probe Reveals Increased Pulmonary Aldehydes in the Injured Newborn Lung
Because we detected increased allysine concentrations in lung specimens obtained from O2-treated mouse pups, we next tested whether aldehyde-reactive compounds could react with them in vivo. It is highly desirable to develop molecular probes that can detect mechanisms of lung injury. This is because they are likely to provide useful tools to analyze specific pathogenic processes causing disease. Moreover, they might be used to guide and follow the efficacy of therapies that are directed at inhibiting or mitigating their role in disrupting pulmonary development and function.
To detect extracellular aldehydes residing in the newborn lung, we utilized a gadolinium-tagged aldehyde-reactive hydrazide-containing compound called Gd-Hyd (Fig. 8A). This compound was designed to be anionic and hydrophilic to minimize nonspecific protein binding, to decrease cellular uptake, and to be rapidly eliminated via renal filtration. We previously showed that it forms reversible covalent bonds with aldehyde groups on proteins and detailed its synthesis, stability, and in vivo distribution (46). We also confirmed that Gd-Hyd selectively binds in vitro with aldehyde-functionalized bovine serum albumin and ex vivo with allysine-rich porcine aortic tissue. In our current study, we determined that O2-mediated lung injury increases in vivo retention of this probe in pulmonary tissues 30 min after it was injected intraperitoneally in the pup. As shown in Fig. 8B, ex vivo measurements of Gd levels in the lung tissues using ICP-MS quantification revealed a 1.5-fold increase in the retention of the aldehyde-reactive probe in the O2-injured pup lung compared with air-breathing control ones. No mouse pup exhibited adverse symptoms associated with treatment with the aldehyde-reactive probe. To our knowledge, this experiment demonstrates the first in vivo utility of a molecular probe in the injured newborn lung.
Figure 8.
O2-mediated injury increases in vivo retention of an aldehyde-reactive probe in the mouse pup lung. A: a gadolinium (Gd)-linked molecular probe (Gd-Hyd) harboring an aldehyde-reactive hydrazide group (circled) was used in experiments to detect in vivo aldehyde levels in the lungs of air- or 85% O2-treated mouse pups. B: O2-induced injury increases the retention of the aldehyde-reactive probe in the mouse pup lungs. Pups treated with either air or 85% O2 were injected with 100 mmol/kg Gd-Hyd intraperitoneally. Thirty minutes later, they were killed and their lungs and soleus and gastrocnemius muscles were obtained. The Gd levels were measured in the tissues using inductively coupled-mass spectroscopy and expressed as nmol Gd/g of lung tissue and as a ratio of the Gd levels detected in the lung and muscle tissue; n = 7/group, *P < 0.05.
Based on this detection of excess in vivo activity of LOXs in the injured newborn lung, we next tested whether the excess aldehydes map to areas where we had detected increased secreted pro-LOX/mature LOX. For this work, we employed an aldehyde-reactive probe that could be detected using immunohistochemistry. DNPH is a compound that contains an aldehyde-reactive hydrazine group covalently linked to a dinitrophenyl ring (Fig. 9A). Previous work has shown that in acidic conditions, DNPH will conjugate with aldehydes, by forming a hydrazone linkage, and that these hydrazones can be detected spectrophotometrically (75). More recently, it was determined that this compound can be detected using an antibody that reacts with its dinitrophenyl ring (76). This approach has been used to localize aldehydes in tissues in models of encephalomyelitis (77). But DNPH also strongly reacts with DNA and with intracellular aldehydes that result from metabolic processes (78). This would decrease the specificity of detecting extracellular aldehydes that might result from secreted LOXs. So we used nonpermeabilized, methacarn-fixed pup lung sections and anti-DNP antibodies to restrict our aldehyde detection to proteins in the extracellular space. As shown in Fig. 9B, using this approach we detected a low level of extracellular aldehydes in the normal mouse pup lung. The background staining of the slide was caused by DNPH reacting with coating substances on it. Attempts to reduce this by performing the assay using uncoated slides led to loss to tissue specimens. Importantly, we identified increased levels of aldehydes using the DNPH molecular probe in the O2-injured pup lung tissues. This was especially evident in interstitial areas that we previously showed harbors increased TGFβ-stimulated myofibroblasts and secreted pro-LOX/mature LOX. The specificity of the antibody is confirmed by its inability to react with lung tissue that has not been treated with DNPH and when the lung aldehydes were reduced using borohydride (NaBH4) before DNPH probe treatment (Fig. 9C). To confirm these results, we also used a different aldehyde reactive compound that relies on a biotin-avidin conjugated enzyme reaction instead of antibody reactivity for retained probe detection. These studies also identified aldehydes primarily in areas where increased secretion of LOX forms had been detected (Supplemental Fig. S6).
Figure 9.
O2-induced injury increases aldehydes in mouse pup lung sections. A: dinitrophenylhydrazine (DNPH) is a molecular probe containing a hydrazine group (circled) bound to a dinitrophenol group (DNP) that reacts with aldehydes forming a hydrazone linkage. The bound compound can then be detected using an anti-DNP antibody and immunohistochemistry (IHC). Performing the assay using nonpermeabilized tissue sections restricts the antibody-mediated aldehyde-reacted probe detection to the extracellular space. B: although a scant level of aldehydes is detected in the air-treated lung, increased levels are observed in the O2-injured mouse pup lung (arrows), particularly in areas previously shown to exhibit increased extracellular pro-LOX and mature LOX and dysmorphic elastin assembly. Methacarn-fixed, nonpermeabilized lungs of mouse pups exposed to air or 85% O2 were reacted DNPH. The retained probe was detected using anti-DNP antibodies, IHC employing a colorimetric substrate (red). The tissue was counterstained with hematoxylin. Typical images of lungs of 8 pups/group are shown. C: the specificity of the aldehyde detection method is demonstrated by decreased DNP detection in O2-injured mouse pup sections not treated with DNPH and those in which the aldehydes were consumed by prior sodium tetrahydroboride (NaBH4) treatment.
LOXs Drive Aldehyde Accumulation in the O2-Injured Mouse Pup Lung
Although the results detailed earlier suggest that extracellular lung aldehydes are increased with O2-mediated injury, the contribution made by LOXs to these levels is unknown. Aldehydes can be formed in tissues during protein degradation and turnover. It is also possible that lung tissue can be oxidized by free radicals associated with breathing gases containing high levels of O2. Certainly, these nonenzymatic mechanisms can play a role in increasing lung aldehyde levels in newborns. For example, previous studies have detected increased levels of aldehyde-containing compounds in the bronchoalveolar fluids of O2-treated animals (79) and newborns (80). So, it is possible that the increased levels of extracellular aldehydes that we detected in the injured mouse pup lung are due to mechanisms that do not involve LOXs. So to determine the direct role of LOXs in increasing aldehydes in the injured newborn lung, we tested whether an inhibitor of LOXs decreases the levels of extracellular aldehydes detected in mouse pup lungs treated with high levels of O2.
BAPN is a potent, irreversible inhibitor of LOXs. In the O2-injured newborn lung, BAPN treatment inhibits the activity of LOXs in bronchoalveolar effluents (39, 81), and improves allysine/lysine crosslinking, ECM formation (39, 81), and the biomechanical properties of the lung (81). To test whether the inhibition of LOXs is also associated with a decrease of pulmonary aldehydes, we treated mouse pups with and without 85% O2 and BAPN and then evaluated the level of extracellular aldehydes using methacarn-fixed pup lung sections, the DNPH probe, and IHC. As shown in Fig. 10A, whereas extracellular aldehydes appeared to be increased in the lung sections of mouse pups with O2-mediated injury, they were decreased in those obtained from pups that were treated with the inhibitor of LOXs. Consistent with the findings of others (39, 81), the abnormal elastin formation caused by O2-mediated injury was improved in the lungs of pups that were treated with the inhibitor of LOXs.
Figure 10.
Lysyl oxidases (LOXs) increase aldehyde levels and disrupt elastogenesis in the O2-injured mouse pup lung. Mouse pups were continuously treated with air or 85% O2 and were treated with or without β-aminopropionitrile (BAPN), an inhibitor of LOXs, as detailed in materials and methods. Subsequently, aldehydes (red) were detected in lung sections using dinitrophenylhydrazine (DNPH) and immunohistochemistry (IHC) with methyl green counterstaining, or elastin morphology was determined using a resorcin fuchsin stain (purple) and tartrazine (yellow) counterstain, as described earlier. In addition to the bright field (380–700 nm wavelength) images, 525 ± 20 nm wavelength images were obtained to select for the aldehydes detected by the IHC staining. A: the increase in aldehydes detected in O2-injured mouse pup lung sections is decreased in those obtained from animals treated with BAPN. Although lung injury causes dysmorphic elastin deposition in alveolar septa, the elastin formation appeared to be improved with BAPN treatment. B: quantification of aldehyde levels indicates that LOX inhibition decreases the level of aldehydes detected in O2-injured mouse pup lung sections. The mean integrated intensity of the aldehyde signal detected in the 525 nm-centered images was determined as described in materials and methods. n = 8/group, *P < 0.008 vs. each of the other treatments; NS, no significant difference.
To quantify the effect of BAPN on the mouse pup lung section aldehyde levels, we adapted an absorptive micro-densitometry method detailed by others (48) and used it to analyze the DNPH-reactive signals in the mouse pup lung sections. This approach takes advantage of the differential wavelength transmittance of select IHC chromogens and counterstains. In particular, it has been noted that whereas the alkaline phosphatase product of VectorRed absorbs nearly all ∼525 nm light, nearly none of light at that wavelength is absorbed by the methyl green counterstain (48). This means that by controlling the tissue thickness, duration of chromogen exposure, and wavelength band of light intensity detection, it is possible to quantify IHC signal using this chromogen and counterstain combination. Studies using precision-cut O2-injured mouse pup lung sections allowed us to determine the chromogen exposure conditions that yielded quantitative resolution of DNPH reactivity (Supplemental Fig. S7). Employing these conditions, we confirmed that whereas BAPN treatment alone caused a negligible decrease in DNPH reactivity in control mouse pup lungs, treatment with this inhibitor of LOXs prevented the elevation in extracellular aldehydes levels in the O2-injured mouse pup lung sections (Fig. 10B). These data demonstrate a direct role of LOXs in generating extracellular aldehydes in the O2-injured mouse pup lung.
DISCUSSION
Accumulating evidence links TGFβ and LOX activation with mechanisms of newborn lung disease. Here, we determined that LOX protein expression is increased in TGFβ-stimulated interstitial fibroblasts in the O2-injured pup lung. The direct role of TGFβ in regulating LOX expression is supported by the determination that TGFβ increases LOX protein detection in primary mouse pup fibroblasts. In embryonic mouse fibroblast cells, TGFβ was also found to increase LOX mRNA and protein expression. Importantly, in this model cell line we determined that TGFβ increases LOX expression by stimulating mRNA transcription via canonical Smad2/3-dependent signaling. LOX undergoes extensive post-translational modifications, which control its secretion and extracellular activation. In the newborn mouse pup, we detected extracellular protein cleavage of pro-LOX forms and release of catalytically active forms in soluble lung fractions. In studying the role of newborn lung injury in regulating LOX expression, we determined that O2-mediated injury increases pro-LOX secretion particularly in areas of dysplastic elastin formation and alveolar development. Using in vivo treatment with novel molecular probes, we determined that the activation of LOXs, such as LOX, in the O2-injured pup increases the levels of extracellular pulmonary aldehydes in the lung. Furthermore, we determined that the extracellular aldehydes map to areas of increased LOX secretion and disrupted ECM and alveolar formation in the pup lung sections. Although it is possible that superoxides and other nonenzymatic processes increase aldehyde generation in the hyperoxic lung tissue, in vivo work with a potent inhibitor of LOXs demonstrated a primary role of the enzymes in stimulating the accumulation of these reactive carbonyl groups in the O2-injured pup lung. Together, these data provide mechanistic information about the processes that dysregulate the expression and activities of LOXs during newborn lung injury.
Previous work indicates that TGFβ increases LOX mRNA expression in fibroblast cell lines and cultured human lung fibroblasts (36). This is confirmed by our demonstration that a similar level of TGFβ that was detected in tracheal aspirates from babies rapidly increases LOX mRNA levels in mouse embryonic NIH3T3 fibroblasts. This treatment was associated also with an increase in LOX protein expression in the NIH3T3 fibroblasts and in the primary mouse pup lung fibroblasts. But, in contrast with reports by others (5), we determined that the TGFβ stimulation did not increase the level of mature LOX in the NIH3T3 fibroblasts. Although we detected protein with a molecular weight consistent with pre-pro-/pro-LOX by immunoblotting, we did not detect one exhibiting a weight of the cleaved forms of the enzyme. Moreover, we did not detect nuclear LOX in the permeabilized NIH3T3 fibroblasts using immunofluorescence. Although it is possible that this represents a failure of the LOX-reacting antibody to penetrate the nucleus of the NIH3T3 fibroblasts, the cells were permeabilized using methods that we previously showed permits nuclear antibody entry (82, 83). Moreover, this method allowed the antibody to react with anti-LOX antibody reactivity in the nuclei of the mouse pup fibroblasts. Although TGFβ increased LOX protein expression in all the cells, the mechanisms that regulate its secretion, cleavage, reuptake, and localization into the nuclei of some but not all cells are unknown at this time.
TGFβ is known to regulate gene expression by controlling mRNA transcription and post-transcriptional mechanisms. Previous studies in cardiac fibroblasts show that TGFβ can regulate mRNA stability (84). In that case, TGFβ was determined to stimulate the localization of HuR from the nucleus to the cytosol of the fibroblasts, the binding of HuR to the 3′UTR of the target RNA, and the stabilization of the mRNA. But, inspection of a database of genes that are predicted to encode RNA that harbors a putative HuR-binding motif (85) did not reveal LOX, although the paralog LOXL-1 was found. Also, in studies in which HuR-mRNA complexes were immunopurified from HeLa cells and sequenced, neither LOX nor LOXL-1 encoding RNA was detected (86). These findings are consistent with our results. This is because in employing transcription inhibition methods, we determined that TGFβ regulates LOX mRNA expression solely at transcriptional level in fibroblasts. During additional work employing RNAi, we determined that TGFβ increases LOX gene expression via canonical Smad2/3- and Smad4-dependent mechanisms. Although this mechanism has not been previously determined in fibroblast cells, others have observed a similar canonical TGFβ regulation mechanism of LOX expression in cultured ocular trabecular meshwork cells (87).
Previous studies suggested that newborn lung injury causes excessive cross-linking of ECM precursor proteins. For example, increased levels of desmosine and dihydroxylysinonorleucine levels were detected in the lung lysates of O2-injured pup lungs (37, 39). These unique amino acids result from the nonenzymatic condensation of allysines and lysine in extracellular proteins; their increased abundance appears to confirm an increased activity of LOXs in this lung injury model. But aldehydes are highly reactive moieties, and it is desirable to determine whether the increase in the activity of LOXs produces levels of aldehydes that are more than that which can be consumed during ECM protein cross-linking. To accomplish this work, we devised several new methods to quantify and map extracellular aldehydes in the O2-injured mouse pup lung. First, we employed a novel and highly sensitive biochemical assay and showed that the total levels of allysine were increased in the injured pup lung. Then, we employed aldehyde-reactive compounds and nonpermeabilized tissues to map extracellular aldehydes in lung sections of mouse pups. These later studies documented an increase in extracellular aldehyde levels in the O2-injured mouse pup lung particularly in areas where we detected increased LOX expression by TGFβ-stimulated interstitial myofibroblasts and dysmorphic elastin assembly. Together, these studies suggest that O2-mediated injury increases the activity of LOXs and extracellular aldehydes in the newborn lung.
Because extracellular aldehydes were increased in the O2-injured mouse pup lung, we tested the exciting possibility that systemic treatment with an engineered molecular probe might detect the activation of LOXs by reacting with excess levels of extracellular aldehydes in vivo. We believed that the aldehyde-reactive Gd-Hyd probe would be tolerated well by the pups because it is rapidly eliminated, is very hydrophilic, exhibits no nonspecific protein binding, is extracellular, and is not metabolized. Previously, we showed that systemic treatments with Gd-Hyd cause no toxicity in adult rodent models of lung and liver disease (88–91). We believed that the increased leakiness of the injured vasculature of the O2-injured pup lung might permit Gd-Hyd to enter the interstitium and detect aldehydes there. This work detailed here showed a high level of retention the aldehyde-reactive probe Gd-Hyd in the O2-injured pup lung.
The increased extracellular aldehydes detected in the injured mouse pup lung did not map precisely to areas where increased secreted LOX was detected. Certainly, it is possible that cells that were transected during lung sectioning revealed intracellular sources of aldehydes that reacted with the DPN-detecting antibody. Also, it is possible that superoxides, inflammatory processes, and other mechanisms might directly increase aldehydes in the extracellular space in areas where LOXs are not active. To test the role of LOXs in the aldehyde generation, we utilized an aldehyde-reactive probe, antibodies, and quantitative absorptive microdensitometry to quantify the levels of aldehydes in pup lungs that were injured with O2 during the treatment with a LOX inhibitor. This work confirmed that LOXs play a role in the increase in extracellular aldehydes.
However, there are some limitations to this work. For example, although we determined that TGFβ stimulates LOX promoter activity and increases LOX mRNA expression via Smad2/3- and Smad4-dependent mechanisms, we did not determine the mechanisms whereby the canonical TGFβ signaling complex regulates the LOX promoter. But, inspection of the murine LOX promoter suggests that it harbors nine or more TGFβ Smad-binding elements (92). Moreover, chromatin immunoprecipitation experiments using mouse osteoblast precursor cell nuclear extracts determined that Smad4 directly binds to and regulates the LOX promoter (92). So it is likely that the binding of Smad2/3-Smad4 complexes to the LOX promoter stimulates its activity. In addition, although we detected anti-LOX immunoreactivity in the nuclei of interstitial and epithelial cells in the O2-injured pup lung, we did not isolate the nuclei from those cells in the tissue and confirm that it harbored the mature form of LOX. Previous studies have detected mature LOX in the nuclei of some cultured cells (5, 70), but its activity there is incompletely understood. Studies show that heterologous expression of LOX causes decondensation of chromatin, and immunoprecipitation experiments have detected the interaction of LOX with histone H1 (for review, see Ref. 61). But a direct role of LOX in regulating nuclear function has not been demonstrated at this point. Recent studies have detected a correlation between nuclear LOX in tissues and pathologic processes. For example, the detection of nuclear LOX in tumor sections was associated with a decreased response to therapeutics and a poor clinical prognosis in patients with high-grade serous ovarian cancer (93). But the direct role of nuclear LOX in the pathogenesis of disease has not been detailed at this time. Another limitation of this work is that we have not studied the pulmonary extracellular aldehyde levels at earlier stages of O2-mediated pup lung injury. But Kumarasamy et al. (37) demonstrated that LOX mRNA and protein levels increase as early as 7 days after the initiation of lung injury in this model. Because our focus was to provide the first link between the activation of LOXs and extracellular aldehyde generation in the O2-induced newborn lung, we focused our efforts studying the regulation after 21 days of lung injury. This is the time shown in the previous work to cause a sustained high level of LOX expression in this model. Based on our novel findings, future studies are warranted to determine time scale of pulmonary TGFβ and LOX activation and extracellular aldehyde accumulation employing the use of molecular probes.
In summary, the work detailed here provides important mechanistic information about the role of TGFβ and LOX in newborn lung injury. They provide a basis for the future development of extracellular aldehyde-sensing biomolecules as tools that might be employed to assess excess activity of LOXs in the injured newborn lung.
ETHICAL APPROVALS
All experiments were approved by the Institutional Animal Care and Use Committee at the MGH.
SUPPLEMENTAL DATA
Supplemental Table S1 and Supplemental Figs. S1–S7: https://doi.org/10.17605/OSF.IO/YDB4V.
GRANTS
This work was supported by the National Heart, Lung, and Blood Institute Grants HL-125715 and HL-147863 (to J. D. Roberts, Jr.), HL-121152 (to H. H. Chen), HL-135242 and HL-151704 (to C. T. Nguyen), and HL-116315 and HL-131907 (to P. Caravan) and the NIH Office of the Director for Instrumentation Grants OD-010650 and OD-025234 (to P. Caravan).
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the authors.
AUTHOR CONTRIBUTIONS
Y.Z., Z.K., H.H.C., P.C., and J.D.R. conceived and designed research; Y.Z., R.C.M., Z.K., H.H.C., and J.D.R. performed experiments; Y.Z., R.C.M., Z.K., H.H.C., C.T.N., P.C., and J.D.R. analyzed data; Y.Z., Z.K., H.H.C., P.C., and J.D.R. interpreted results of experiments; Y.Z., Z.K., H.H.C., and J.D.R. prepared figures; Y.Z., Z.K., H.H.C., and J.D.R. drafted manuscript; Y.Z., R.C.M., Z.K., H.H.C., C.T.N., P.C., and J.D.R. edited and revised manuscript; Y.Z., R.C.M., Z.K., H.H.C., C.T.N., P.C., and J.D.R. approved final version of manuscript.
ACKNOWLEDGMENTS
We thank Eman Akam, Matthew Drummond, and Timothy Robbins for technical help, Kathleen Flanders and Lalage Wakefield (National Cancer Institute) for providing key reagents, and Chelin Hu and Oscar Bizzozero (Univ. of New Mexico) for giving detailed advice about the DNPH aldehyde assay.
Present address of Z. Khatun: Dept. of Pharmaceutical Sciences, University of Texas at El Paso, El Paso, TX.
REFERENCES
- 1.Csiszar K. Lysyl oxidases: a novel multifunctional amine oxidase family. Prog Nucleic Acid Res Mol Biol 70: 1–32, 2001. doi: 10.1016/s0079-6603(01)70012-8. [DOI] [PubMed] [Google Scholar]
- 2.Kagan HM, Li W. Lysyl oxidase: properties, specificity, and biological roles inside and outside of the cell. J Cell Biochem 88: 660–672, 2003. doi: 10.1002/jcb.10413. [DOI] [PubMed] [Google Scholar]
- 3.Kozel BA, Wachi H, Davis EC, Mecham RP. Domains in tropoelastin that mediate elastin deposition in vitro and in vivo. J Biol Chem 278: 18491–18498, 2003. doi: 10.1074/jbc.M212715200. [DOI] [PubMed] [Google Scholar]
- 4.Lucero HA, Ravid K, Grimsby JL, Rich CB, DiCamillo SJ, Maki JM, Myllyharju J, Kagan HM. Lysyl oxidase oxidizes cell membrane proteins and enhances the chemotactic response of vascular smooth muscle cells. J Biol Chem 283: 24103–24117, 2008. doi: 10.1074/jbc.M709897200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Li W, Nellaiappan K, Strassmaier T, Graham L, Thomas KM, Kagan HM. Localization and activity of lysyl oxidase within nuclei of fibrogenic cells. Proc Natl Acad Sci USA 94: 12817–12822, 1997. doi: 10.1073/pnas.94.24.12817. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Giampuzzi M, Botti G, Di Duca M, Arata L, Ghiggeri G, Gusmano R, Ravazzolo R, Di Donato A. Lysyl oxidase activates the transcription activity of human collagene III promoter. Possible involvement of Ku antigen. J Biol Chem 275: 36341–36349, 2000. doi: 10.1074/jbc.M003362200. [DOI] [PubMed] [Google Scholar]
- 7.Mello ML, Contente S, Vidal BC, Planding W, Schenck U. Modulation of ras transformation affecting chromatin supraorganization as assessed by image analysis. Exp Cell Res 220: 374–382, 1995. doi: 10.1006/excr.1995.1328. [DOI] [PubMed] [Google Scholar]
- 8.Mizikova I, Palumbo F, Tabi T, Herold S, Vadasz I, Mayer K, Seeger W, Morty RE. Perturbations to lysyl oxidase expression broadly influence the transcriptome of lung fibroblasts. Physiol Genomics 49: 416–429, 2017. doi: 10.1152/physiolgenomics.00026.2017. [DOI] [PubMed] [Google Scholar]
- 9.Oleggini R, Gastaldo N, Di Donato A. Regulation of elastin promoter by lysyl oxidase and growth factors: cross control of lysyl oxidase on TGF-β1 effects. Matrix Biol 26: 494–505, 2007. doi: 10.1016/j.matbio.2007.02.003. [DOI] [PubMed] [Google Scholar]
- 10.Myllyla R, Wang C, Heikkinen J, Juffer A, Lampela O, Risteli M, Ruotsalainen H, Salo A, Sipila L. Expanding the lysyl hydroxylase toolbox: new insights into the localization and activities of lysyl hydroxylase 3 (LH3). J Cell Physiol 212: 323–329, 2007. doi: 10.1002/jcp.21036. [DOI] [PubMed] [Google Scholar]
- 11.Salo AM, Wang C, Sipila L, Sormunen R, Vapola M, Kervinen P, Ruotsalainen H, Heikkinen J, Myllyla R. Lysyl hydroxylase 3 (LH3) modifies proteins in the extracellular space, a novel mechanism for matrix remodeling. J Cell Physiol 207: 644–653, 2006. doi: 10.1002/jcp.20596. [DOI] [PubMed] [Google Scholar]
- 12.Mizikova I, Pfeffer T, Nardiello C, Surate Solaligue DE, Steenbock H, Tatsukawa H, Silva DM, Vadasz I, Herold S, Pease RJ, Iismaa SE, Hitomi K, Seeger W, Brinckmann J, Morty RE. Targeting transglutaminase 2 partially restores extracellular matrix structure but not alveolar architecture in experimental bronchopulmonary dysplasia. FEBS J 285: 3056–3076, 2018. doi: 10.1111/febs.14596. [DOI] [PubMed] [Google Scholar]
- 13.Schittny JC, Paulsson M, Vallan C, Burri PH, Kedei N, Aeschlimann D. Protein cross-linking mediated by tissue transglutaminase correlates with the maturation of extracellular matrices during lung development. Am J Respir Cell Mol Biol 17: 334–343, 1997. doi: 10.1165/ajrcmb.17.3.2737. [DOI] [PubMed] [Google Scholar]
- 14.Leask A, Abraham DJ. TGF-β signaling and the fibrotic response. FASEB J 18: 816–827, 2004. doi: 10.1096/fj.03-1273rev. [DOI] [PubMed] [Google Scholar]
- 15.Bland RD, Ertsey R, Mokres LM, Xu L, Jacobson BE, Jiang S, Alvira CM, Rabinovitch M, Shinwell ES, Dixit A. Mechanical ventilation uncouples synthesis and assembly of elastin and increases apoptosis in lungs of newborn mice. Prelude to defective alveolar septation during lung development? Am J Physiol Lung Cell Mol Physiol 294: L3–L14, 2008. doi: 10.1152/ajplung.00362.2007. [DOI] [PubMed] [Google Scholar]
- 16.Nakanishi H, Sugiura T, Streisand JB, Lonning SM, Roberts JD Jr.. TGF-β-neutralizing antibodies improve pulmonary alveologenesis and vasculogenesis in the injured newborn lung. Am J Physiol Lung Cell Mol Physiol 293: L151–L161, 2007. doi: 10.1152/ajplung.00389.2006. [DOI] [PubMed] [Google Scholar]
- 17.Albertine KH, Jones GP, Starcher BC, Bohnsack JF, Davis PL, Cho SC, Carlton DP, Bland RD. Chronic lung injury in preterm lambs. Disordered respiratory tract development. Am J Respir Crit Care Med 159: 945–958, 1999. doi: 10.1164/ajrccm.159.3.9804027. [DOI] [PubMed] [Google Scholar]
- 18.Bland RD, Albertine KH, Pierce RA, Starcher BC, Carlton DP. Impaired alveolar development and abnormal lung elastin in preterm lambs with chronic lung injury: potential benefits of retinol treatment. Biol Neonate 84: 101–102, 2003. doi: 10.1159/000071012. [DOI] [PubMed] [Google Scholar]
- 19.Bland RD, Xu L, Ertsey R, Rabinovitch M, Albertine KH, Wynn KA, Kumar V, Ryan RM, Swartz DD, Csiszar K, Fong KS. Dysregulation of pulmonary elastin synthesis and assembly in preterm lambs with chronic lung disease. Am J Physiol Lung Cell Mol Physiol 292: L1370–L1384 2007. doi: 10.1152/ajplung.00367.2006. [DOI] [PubMed] [Google Scholar]
- 20.Pierce RA, Albertine KH, Starcher BC, Bohnsack JF, Carlton DP, Bland RD. Chronic lung injury in preterm lambs: disordered pulmonary elastin deposition. Am J Physiol Lung Cell Mol Physiol 272: L452–L460, 1997. doi: 10.1152/ajplung.1997.272.3.L452. [DOI] [PubMed] [Google Scholar]
- 21.Coalson JJ, Winter VT, Siler-Khodr T, Yoder BA. Neonatal chronic lung disease in extremely immature baboons. Am J Respir Crit Care Med 160: 1333–1346, 1999. doi: 10.1164/ajrccm.160.4.9810071. [DOI] [PubMed] [Google Scholar]
- 22.Bruce MC, Wedig KE, Jentoft N, Martin RJ, Cheng PW, Boat TF, Fanaroff AA. Altered urinary excretion of elastin cross-links in premature infants who develop bronchopulmonary dysplasia. Am Rev Respir Dis 131: 568–572, 1985. doi: 10.1164/arrd.1985.131.4.568. [DOI] [PubMed] [Google Scholar]
- 23.Hislop AA, Wigglesworth JS, Desai R, Aber V. The effects of preterm delivery and mechanical ventilation on human lung growth. Early Hum Dev 15: 147–164, 1987. doi: 10.1016/0378-3782(87)90003-x. [DOI] [PubMed] [Google Scholar]
- 24.Thibeault DW, Mabry SM, Ekekezie II, Truog WE. Lung elastic tissue maturation and perturbations during the evolution of chronic lung disease. Pediatrics 106: 1452–1459, 2000. doi: 10.1542/peds.106.6.1452. [DOI] [PubMed] [Google Scholar]
- 25.Thibeault DW, Mabry SM, Ekekezie II, Zhang X, Truog WE. Collagen scaffolding during development and its deformation with chronic lung disease. Pediatrics 111: 766–776, 2003. doi: 10.1542/peds.111.4.766. [DOI] [PubMed] [Google Scholar]
- 26.Alejandre-Alcazar MA, Kwapiszewska G, Reiss I, Amarie OV, Marsh LM, Sevilla-Perez J, Wygrecka M, Eul B, Koebrich S, Hesse M, Schermuly RT, Seeger W, Eickelberg O, Morty RE. Hyperoxia modulates TGF-beta/BMP signaling in a mouse model of bronchopulmonary dysplasia. Am J Physiol Lung Cell Mol Physiol 292: L537–L549, 2007. doi: 10.1152/ajplung.00050.2006. [DOI] [PubMed] [Google Scholar]
- 27.Buckley S, Warburton D. Dynamics of metalloproteinase-2 and -9, TGF-β, and uPA activities during normoxic vs. hyperoxic alveolarization. Am J Physiol Lung Cell Mol Physiol 283: L747–L754, 2002. doi: 10.1152/ajplung.00415.2001. [DOI] [PubMed] [Google Scholar]
- 28.Dasgupta C, Sakurai R, Wang Y, Guo P, Ambalavanan N, Torday JS, Rehan VK. Hyperoxia-induced neonatal rat lung injury involves activation of TGF-β and Wnt signaling and is protected by rosiglitazone. Am J Physiol Lung Cell Mol Physiol 296: L1031–L1041, 2009. doi: 10.1152/ajplung.90392.2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Liao J, Kapadia VS, Brown LS, Cheong N, Longoria C, Mija D, Ramgopal M, Mirpuri J, McCurnin DC, Savani RC. The NLRP3 inflammasome is critically involved in the development of bronchopulmonary dysplasia. Nat Commun 6: 8977, 2015. doi: 10.1038/ncomms9977. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ambalavanan N, Nicola T, Hagood J, Bulger A, Serra R, Murphy-Ullrich J, Oparil S, Chen YF. Transforming growth factor-beta signaling mediates hypoxia-induced pulmonary arterial remodeling and inhibition of alveolar development in newborn mouse lung. Am J Physiol Lung Cell Mol Physiol 295: L86–L95, 2008. doi: 10.1152/ajplung.00534.2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Vicencio AG, Eickelberg O, Stankewich MC, Kashgarian M, Haddad GG. Regulation of TGF-beta ligand and receptor expression in neonatal rat lungs exposed to chronic hypoxia. J Appl Physiol (1985) 93: 1123–1130, 2002. doi: 10.1152/japplphysiol.00031.2002. [DOI] [PubMed] [Google Scholar]
- 32.Albertine KH, MacRitchie AN, Young BJ, Cho SC, Kullama LK, Carlton DP, Bland RD. Altered vascular development in preterm lambs with chronic lung injury. Chest 114: 6S–7S, 1998. doi: 10.1378/chest.114.1_supplement.6s-a. [DOI] [PubMed] [Google Scholar]
- 33.Hilgendorff A, Parai K, Ertsey R, Jain N, Navarro EF, Peterson JL, Tamosiuniene R, Nicolls MR, Starcher BC, Rabinovitch M, Bland RD. Inhibiting lung elastase activity enables lung growth in mechanically ventilated newborn mice. Am J Respir Crit Care Med 184: 537–546, 2011. doi: 10.1164/rccm.201012-2010OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Mokres LM, Parai K, Hilgendorff A, Ertsey R, Alvira CM, Rabinovitch M, Bland RD. Prolonged mechanical ventilation with air induces apoptosis and causes failure of alveolar septation and angiogenesis in lungs of newborn mice. Am J Physiol Lung Cell Mol Physiol 298: L23–L35, 2010. doi: 10.1152/ajplung.00251.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Boak AM, Roy R, Berk J, Taylor L, Polgar P, Goldstein RH, Kagan HM. Regulation of lysyl oxidase expression in lung fibroblasts by transforming growth factor-β 1 and prostaglandin E2. Am J Respir Cell Mol Biol 11: 751–755, 1994. doi: 10.1165/ajrcmb.11.6.7946403. [DOI] [PubMed] [Google Scholar]
- 36.Aumiller V, Strobel B, Romeike M, Schuler M, Stierstorfer BE, Kreuz S. Comparative analysis of lysyl oxidase (like) family members in pulmonary fibrosis. Sci Rep 7: 149, 2017. doi: 10.1038/s41598-017-00270-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Kumarasamy A, Schmitt I, Nave AH, Reiss I, van der Horst I, Dony E, Roberts JD Jr, de Krijger RR, Tibboel D, Seeger W, Schermuly RT, Eickelberg O, Morty RE. Lysyl oxidase activity is dysregulated during impaired alveolarization of mouse and human lungs. Am J Respir Crit Care Med 180: 1239–1252, 2009. doi: 10.1164/rccm.200902-0215OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Witsch TJ, Turowski P, Sakkas E, Niess G, Becker S, Herold S, Mayer K, Vadasz I, Roberts JD Jr, Seeger W, Morty RE. Deregulation of the lysyl hydroxylase matrix cross-linking system in experimental and clinical bronchopulmonary dysplasia. Am J Physiol Lung Cell Mol Physiol 306: L246–L259, 2014. doi: 10.1152/ajplung.00109.2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Mizikova I, Ruiz-Camp J, Steenbock H, Madurga A, Vadasz I, Herold S, Mayer K, Seeger W, Brinckmann J, Morty RE. Collagen and elastin cross-linking is altered during aberrant late lung development associated with hyperoxia. Am J Physiol Lung Cell Mol Physiol 308: L1145–L1158, 2015. doi: 10.1152/ajplung.00039.2015. [DOI] [PubMed] [Google Scholar]
- 40.Pozarska A, Rodriguez-Castillo JA, Surate Solaligue DE, Ntokou A, Rath P, Mizikova I, Madurga A, Mayer K, Vadasz I, Herold S, Ahlbrecht K, Seeger W, Morty RE. Stereological monitoring of mouse lung alveolarization from the early postnatal period to adulthood. Am J Physiol Lung Cell Mol Physiol 312: L882–L895, 2017. doi: 10.1152/ajplung.00492.2016. [DOI] [PubMed] [Google Scholar]
- 41.Branchfield K, Li R, Lungova V, Verheyden JM, McCulley D, Sun X. A three-dimensional study of alveologenesis in mouse lung. Dev Biol 409: 429–441, 2016. doi: 10.1016/j.ydbio.2015.11.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Turner N, Lim XY, Toop HD, Osborne B, Brandon AE, Taylor EN, Fiveash CE, Govindaraju H, Teo JD, McEwen HP, Couttas TA, Butler SM, Das A, Kowalski GM, Bruce CR, Hoehn KL, Fath T, Schmitz-Peiffer C, Cooney GJ, Montgomery MK, Morris JC, Don AS. A selective inhibitor of ceramide synthase 1 reveals a novel role in fat metabolism. Nat Commun 9: 3165, 2018. doi: 10.1038/s41467-018-05613-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Forster B, Van De Ville D, Berent J, Sage D, Unser M. Complex wavelets for extended depth-of-field: a new method for the fusion of multichannel microscopy images. Microsc Res Tech 65: 33–42, 2004. doi: 10.1002/jemt.20092. [DOI] [PubMed] [Google Scholar]
- 44.Elkouris M, Kontaki H, Stavropoulos A, Antonoglou A, Nikolaou KC, Samiotaki M, Szantai E, Saviolaki D, Brown PJ, Sideras P, Panayotou G, Talianidis I. SET9-mediated regulation of TGF-β signaling links protein methylation to pulmonary fibrosis. Cell Rep 15: 2733–2744, 2016. doi: 10.1016/j.celrep.2016.05.051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Contente S, Csiszar K, Kenyon K, Friedman RM. Structure of the mouse lysyl oxidase gene. Genomics 16: 395–400, 1993. doi: 10.1006/geno.1993.1202. [DOI] [PubMed] [Google Scholar]
- 46.Chen HH, Waghorn PA, Wei L, Tapias LF, Schu Hle DT, Rotile NJ, Jones CM, Looby RJ, Zhao G, Elliott JM, Probst CK, Mino-Kenudson M, Lauwers GY, Tager AM, Tanabe KK, Lanuti M, Fuchs BC, Caravan P. Molecular imaging of oxidized collagen quantifies pulmonary and hepatic fibrogenesis. JCI insight 2: e91506, 2017. doi: 10.1172/jci.insight.91506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Waghorn PA, Oliveira BL, Jones CM, Tager AM, Caravan P. High sensitivity HPLC method for determination of the allysine concentration in tissue by use of a naphthol derivative. J Chromatogr B Analyt Technol Biomed Life Sci 1064: 7–13, 2017. doi: 10.1016/j.jchromb.2017.08.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Ermert L, Hocke AC, Duncker HR, Seeger W, Ermert M. Comparison of different detection methods in quantitative microdensitometry. Am J Pathol 158: 407–417, 2001. doi: 10.1016/S0002-9440(10)63984-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Akam EA, Abston E, Rotile NJ, Slattery HR, Zhou IY, Lanuti M, Caravan P. Improving the reactivity of hydrazine-bearing MRI probes for in vivo imaging of lung fibrogenesis. Chem Sci 11: 224–231, 2020. doi: 10.1039/c9sc04821a. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing, 2018. https://www.r-project.org/. [Google Scholar]
- 51.Gu L, Zhu YJ, Yang X, Guo ZJ, Xu WB, Tian XL. Effect of TGF-β/Smad signaling pathway on lung myofibroblast differentiation. Acta Pharmacol Sin 28: 382–391, 2007. doi: 10.1111/j.1745-7254.2007.00468.x. [DOI] [PubMed] [Google Scholar]
- 52.Hashimoto S, Gon Y, Takeshita I, Matsumoto K, Maruoka S, Horie T. Transforming growth Factor-beta1 induces phenotypic modulation of human lung fibroblasts to myofibroblast through a c-Jun-NH2-terminal kinase-dependent pathway. Am J Respir Crit Care Med 163: 152–157, 2001. doi: 10.1164/ajrccm.163.1.2005069. [DOI] [PubMed] [Google Scholar]
- 53.Kimani PW, Holmes AJ, Grossmann RE, McGowan SE. PDGF-Rα gene expression predicts proliferation, but PDGF-A suppresses transdifferentiation of neonatal mouse lung myofibroblasts. Respir Res 10: 119, 2009. doi: 10.1186/1465-9921-10-119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54.Jourdan-Le Saux C, Gleyzal C, Raccurt M, Sommer P. Functional analysis of the lysyl oxidase promoter in myofibroblast-like clones of 3T6 fibroblast. J Cell Biochem 64: 328–341, 1997. doi:. [DOI] [PubMed] [Google Scholar]
- 55.Desmouliere A, Geinoz A, Gabbiani F, Gabbiani G. Transforming growth factor-beta 1 induces alpha-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J Cell Biol 122: 103–111, 1993. doi: 10.1083/jcb.122.1.103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Roy SG, Nozaki Y, Phan SH. Regulation of α-smooth muscle actin gene expression in myofibroblast differentiation from rat lung fibroblasts. Int J Biochem Cell Biol 33: 723–734, 2001. doi: 10.1016/s1357-2725(01)00041-3. [DOI] [PubMed] [Google Scholar]
- 57.Endale M, Ahlfeld S, Bao E, Chen X, Green J, Bess Z, Weirauch MT, Xu Y, Perl AK. Temporal, spatial, and phenotypical changes of PDGFRα expressing fibroblasts during late lung development. Dev Biol 425: 161–175, 2017. doi: 10.1016/j.ydbio.2017.03.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Masur SK, Dewal HS, Dinh TT, Erenburg I, Petridou S. Myofibroblasts differentiate from fibroblasts when plated at low density. Proc Natl Acad Sci USA 93: 4219–4223, 1996. doi: 10.1073/pnas.93.9.4219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Kotecha S, Wangoo A, Silverman M, Shaw RJ. Increase in the concentration of transforming growth factor beta-1 in bronchoalveolar lavage fluid before development of chronic lung disease of prematurity. J Pediatr 128: 464–469, 1996. doi: 10.1016/s0022-3476(96)70355-4. [DOI] [PubMed] [Google Scholar]
- 60.Hata A, Chen YG. TGF-β signaling from receptors to Smads. Cold Spring Harb Perspect Biol 8: a022061, 2016. doi: 10.1101/cshperspect.a022061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Iturbide A, Garcia de Herreros A, Peiro S. A new role for LOX and LOXL2 proteins in transcription regulation. FEBS J 282: 1768–1773, 2015. doi: 10.1111/febs.12961. [DOI] [PubMed] [Google Scholar]
- 62.Gacheru SN, Thomas KM, Murray SA, Csiszar K, Smith-Mungo LI, Kagan HM. Transcriptional and post-transcriptional control of lysyl oxidase expression in vascular smooth muscle cells: effects of TGF-beta 1 and serum deprivation. J Cell Biochem 65: 395–407, 1997. doi:. [DOI] [PubMed] [Google Scholar]
- 63.Csiszar K, Entersz I, Trackman PC, Samid D, Boyd CD. Functional analysis of the promoter and first intron of the human lysyl oxidase gene. Mol Biol Rep 23: 97–108, 1996. doi: 10.1007/BF00424435. [DOI] [PubMed] [Google Scholar]
- 64.Gao S, Zhao Y, Kong L, Toselli P, Chou IN, Stone P, Li W. Cloning and characterization of the rat lysyl oxidase gene promoter: identification of core promoter elements and functional nuclear factor I-binding sites. J Biol Chem 282: 25322–25337, 2007. doi: 10.1074/jbc.M610108200. [DOI] [PubMed] [Google Scholar]
- 65.Zhang YE. Non-Smad signaling pathways of the TGF-β family. Cold Spring Harb Perspect Biol 9: a022129, 2017. doi: 10.1101/cshperspect.a022129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 66.Du L, Roberts JD Jr.. Transforming growth factor-beta down-regulates sGC subunit expression in pulmonary artery smooth muscle cells via MEK and ERK signaling. Am J Physiol Lung Cell Mol Physiol 316: L20–L34, 2019. doi: 10.1152/ajplung.00319.2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Zhang H, Du L, Zhong Y, Flanders KC, Roberts JD. Jr.,Transforming growth factor-beta stimulates Smad1/5 signaling in pulmonary artery smooth muscle cells and fibroblasts of the newborn mouse through ALK1. Am J Physiol Lung Cell Mol Physiol 313: L615–L627, 2017. doi: 10.1152/ajplung.00079.2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 68.Hornstra IK, Birge S, Starcher B, Bailey AJ, Mecham RP, Shapiro SD. Lysyl oxidase is required for vascular and diaphragmatic development in mice. J Biol Chem 278: 14387–14393, 2003. doi: 10.1074/jbc.M210144200. [DOI] [PubMed] [Google Scholar]
- 69.Maki JM, Sormunen R, Lippo S, Kaarteenaho-Wiik R, Soininen R, Myllyharju J. Lysyl oxidase is essential for normal development and function of the respiratory system and for the integrity of elastic and collagen fibers in various tissues. Am J Pathol 167: 927–936, 2005. doi: 10.1016/S0002-9440(10)61183-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 70.Nellaiappan K, Risitano A, Liu G, Nicklas G, Kagan HM. Fully processed lysyl oxidase catalyst translocates from the extracellular space into nuclei of aortic smooth-muscle cells. J Cell Biochem 79: 576–582, 2000. doi:. [DOI] [PubMed] [Google Scholar]
- 71.Hayashi K, Fong KS, Mercier F, Boyd CD, Csiszar K, Hayashi M. Comparative immunocytochemical localization of lysyl oxidase (LOX) and the lysyl oxidase-like (LOXL) proteins: changes in the expression of LOXL during development and growth of mouse tissues. J Mol Histol 35: 845–855, 2004. doi: 10.1007/s10735-004-2340-1. [DOI] [PubMed] [Google Scholar]
- 72.Mamo LB, Suliman HB, Giles BL, Auten RL, Piantadosi CA, Nozik-Grayck E. Discordant extracellular superoxide dismutase expression and activity in neonatal hyperoxic lung. Am J Respir Crit Care Med 170: 313–318, 2004. doi: 10.1164/rccm.200309-1282OC. [DOI] [PubMed] [Google Scholar]
- 73.Ratner V, Slinko S, Utkina-Sosunova I, Starkov A, Polin RA, Ten VS. Hypoxic stress exacerbates hyperoxia-induced lung injury in a neonatal mouse model of bronchopulmonary dysplasia. Neonatology 95: 299–305, 2009. doi: 10.1159/000178798. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 74.Uchida K. Role of reactive aldehyde in cardiovascular diseases. Free Radic Biol Med 28: 1685–1696, 2000. doi: 10.1016/s0891-5849(00)00226-4. [DOI] [PubMed] [Google Scholar]
- 75.Levine RL, Wehr N, Williams JA, Stadtman ER, Shacter E. Determination of carbonyl groups in oxidized proteins. Methods Mol Biol 99: 15–24, 2000. doi: 10.1385/1-59259-054-3:15. [DOI] [PubMed] [Google Scholar]
- 76.Robinson CE, Keshavarzian A, Pasco DS, Frommel TO, Winship DH, Holmes EW. Determination of protein carbonyl groups by immunoblotting. Anal Biochem 266: 48–57, 1999. doi: 10.1006/abio.1998.2932. [DOI] [PubMed] [Google Scholar]
- 77.Zheng J, Bizzozero OA. Accumulation of protein carbonyls within cerebellar astrocytes in murine experimental autoimmune encephalomyelitis. J Neurosci Res 88: 3376–3385, 2010. doi: 10.1002/jnr.22488. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 78.Luo S, Wehr NB. Protein carbonylation: avoiding pitfalls in the 2,4-dinitrophenylhydrazine assay. Redox Rep 14: 159–166, 2009. doi: 10.1179/135100009X392601. [DOI] [PubMed] [Google Scholar]
- 79.Lenz AG, Costabel U, Shaltiel S, Levine RL. Determination of carbonyl groups in oxidatively modified proteins by reduction with tritiated sodium borohydride. Anal Biochem 177: 419–425, 1989. doi: 10.1016/0003-2697(89)90077-8. [DOI] [PubMed] [Google Scholar]
- 80.Gladstone IM Jr, Levine RL. Oxidation of proteins in neonatal lungs. Pediatrics 93: 764–768, 1994. [PubMed] [Google Scholar]
- 81.Mammoto T, Jiang E, Jiang A, Mammoto A. Extracellular matrix structure and tissue stiffness control postnatal lung development through the lipoprotein receptor-related protein 5/Tie2 signaling system. Am J Respir Cell Mol Biol 49: 1009–1018, 2013. doi: 10.1165/rcmb.2013-0147OC. [DOI] [PubMed] [Google Scholar]
- 82.Chen J, Roberts JD Jr.. cGMP-dependent protein kinase I γ encodes a nuclear localization signal that regulates nuclear compartmentation and function. Cell Signal 26: 2633–2644, 2014. doi: 10.1016/j.cellsig.2014.08.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 83.Sugiura T, Nakanishi H, Roberts JD Jr.. Proteolytic processing of cGMP-dependent protein kinase I mediates nuclear cGMP signaling in vascular smooth muscle cells. Circ Res 103: 53–60, 2008. doi: 10.1161/CIRCRESAHA.108.176321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 84.Bai D, Gao Q, Li C, Ge L, Gao Y, Wang H. A conserved TGFβ1/HuR feedback circuit regulates the fibrogenic response in fibroblasts. Cell Signal 24: 1426–1432, 2012. doi: 10.1016/j.cellsig.2012.03.003. [DOI] [PubMed] [Google Scholar]
- 85.Lopez de Silanes I, Zhan M, Lal A, Yang X, Gorospe M. Identification of a target RNA motif for RNA-binding protein HuR. Proc Natl Acad Sci USA 101: 2987–2992, 2004. doi: 10.1073/pnas.0306453101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 86.Lebedeva S, Jens M, Theil K, Schwanhausser B, Selbach M, Landthaler M, Rajewsky N. Transcriptome-wide analysis of regulatory interactions of the RNA-binding protein HuR. Mol Cell 43: 340–352, 2011. doi: 10.1016/j.molcel.2011.06.008. [DOI] [PubMed] [Google Scholar]
- 87.Sethi A, Mao W, Wordinger RJ, Clark AF. Transforming growth factor-beta induces extracellular matrix protein cross-linking lysyl oxidase (LOX) genes in human trabecular meshwork cells. Invest Ophthalmol Vis Sci 52: 5240–5250, 2011. doi: 10.1167/iovs.11-7287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 88.Dos Santos Ferreira D, Arora G, Gieseck RL 3rd, Rotile NJ, Waghorn PA, Tanabe KK, Wynn TA, Caravan P, Fuchs BC. Molecular magnetic resonance imaging of liver fibrosis and fibrogenesis is not altered by inflammation. Invest Radiol 56: 244–251, 2021. doi: 10.1097/RLI.0000000000000737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 89.Erstad DJ, Farrar CT, Ghoshal S, Masia R, Ferreira DS, Chen YI, Choi JK, Wei L, Waghorn PA, Rotile NJ, Tu C, Graham-O'Regan KA, Sojoodi M, Li S, Li Y, Wang G, Corey KE, Or YS, Jiang L, Tanabe KK, Caravan P, Fuchs BC. Molecular magnetic resonance imaging accurately measures the antifibrotic effect of EDP-305, a novel farnesoid X receptor agonist. Hepatol Commun 2: 821–835, 2018. doi: 10.1002/hep4.1193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 90.Waghorn PA, Ferreira DS, Erstad DJ, Rotile NJ, Masia R, Jones CM, Tu C, Sojoodi M, Chen YI, Schlerman F, Wellen J, Martinez RVP, Tanabe KK, Fuchs BC, Caravan P. Quantitative, noninvasive MRI characterization of disease progression in a mouse model of non-alcoholic steatohepatitis. Sci Rep 11: 6105, 2021. [Erratum in Sci Rep 11: 18167, 2021]. doi: 10.1038/s41598-021-85679-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 91.Zhou IY, Clavijo Jordan V, Rotile NJ, Akam E, Krishnan S, Arora G, Krishnan H, Slattery H, Warner N, Mercaldo N, Farrar CT, Wellen J, Martinez R, Schlerman F, Tanabe KK, Fuchs BC, Caravan P. Advanced MRI of liver fibrosis and treatment response in a rat model of nonalcoholic steatohepatitis. Radiology 296: 67–75, 2020. doi: 10.1148/radiol.2020192118. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 92.Salazar VS, Zarkadis N, Huang L, Norris J, Grimston SK, Mbalaviele G, Civitelli R. Embryonic ablation of osteoblast Smad4 interrupts matrix synthesis in response to canonical Wnt signaling and causes an osteogenesis-imperfecta-like phenotype. J Cell Sci 126: 4974–4984, 2013. doi: 10.1242/jcs.131953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 93.De Donato M, Petrillo M, Martinelli E, Filippetti F, Zannoni GF, Scambia G, Gallo D. Uncovering the role of nuclear lysyl oxidase (LOX) in advanced high grade serous ovarian cancer. Gynecol Oncol 146: 170–178, 2017. doi: 10.1016/j.ygyno.2017.05.001. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Table S1 and Supplemental Figs. S1–S7: https://doi.org/10.17605/OSF.IO/YDB4V.