RESEARCH ARTICLE Translational Research in Acute Lung Injury and

Pulmonary Fibrosis

Impact of ventilation-induced lung injury on the structure and function oflamellar bodies

Scott Milos,1,2* Reza Khazaee,1,2* Lynda A. McCaig,1 Karen Nygard,5 Richard B. Gardiner,4,5 Yi Y. Zuo,6

Cory Yamashita,1,2,3 and Ruud Veldhuizen1,2,3

1Lawson Health Research Institute, Western University, London Ontario, Canada; 2Department of Physiology andPharmacology, Western University, London Ontario, Canada; 3Department of Medicine, Western University, London Ontario,Canada; 4Department of Biology, Western University, London Ontario, Canada; 5Biotron Research Centre, WesternUniversity, London Ontario, Canada; and 6Department of Mechanical Engineering, University of Hawaii at Manoa,Honolulu, Hawaii

Submitted 9 February 2017; accepted in final form 17 May 2017

Milos S, Khazaee R, McCaig LA, Nygard K, Gardiner RB, ZuoYY, Yamashita C, Veldhuizen R. Impact of ventilation-induced lunginjury on the structure and function of lamellar bodies. Am J PhysiolLung Cell Mol Physiol 313: L524–L533, 2017. First published May25, 2017; doi:10.1152/ajplung.00055.2017.—Alterations to the pul-monary surfactant system have been observed consistently in venti-lation-induced lung injury (VILI) including composition changes andimpairments in the surface tension reducing ability of the isolatedextracellular surfactant. However, there is limited information aboutthe effects of VILI on the intracellular form of surfactant, the lamellarbody. It is hypothesized that VILI leads to alterations of lamellar bodynumbers and function. To test this hypothesis, rats were randomizedto one of three groups, nonventilated controls, control ventilation, andhigh tidal volume ventilation (VILI). Following physiological assess-ment to confirm lung injury, isolated lamellar bodies were tested forsurfactant function on a constrained sessile drop surfactometer. Aseparate cohort of animals was used to fix the lungs followed byexamination of lamellar body numbers and morphology using trans-mission electron microscopy. The results showed an impaired abilityof reducing surface tension for the lamellar bodies isolated from theVILI group as compared with the two other groups. The morpholog-ical assessment revealed that the number, and the relative area coveredby, lamellar bodies were significantly decreased in animals with VILIanimals as compared with the other groups. It is concluded that VILIcauses significant alterations to lamellar bodies. It is speculated thatincreased secretion causes a depletion of lamellar bodies that cannotbe compensated by de novo synthesis of surfactant in these injuredlungs.

surfactant; lamellar bodies; mechanical ventilation; lung injury; bio-physical activity; transmission electron microscopy

MECHANICAL VENTILATION is a supportive therapy used in avariety of clinical settings to maintain adequate gas exchangein patients (2). For people with an underlying lung injury, suchas patients with acute respiratory distress syndrome (ARDS),mechanical ventilation is essential but can also propagate thelung dysfunction and thereby contribute to disease progression,

as is evident by decreasing levels of blood oxygenation duringventilation (35). These damaging effects of mechanical venti-lation are due to repeated opening and collapse at low lungvolumes and overdistention of the lung due to the deliveredtidal volumes in a heterogeneously injured lung (24). Althoughventilation strategies attempting to mitigate these adverseevents have improved outcomes in ARDS, avoiding the me-chanical ventilation-induced damaged completely is not feasi-ble (5, 37). To further improve outcomes of ARDS patients, itis important to understand the mechanisms by which mechan-ical ventilation affects the lung.

Animal models, in which the injury is termed ventilation-induced lung injury (VILI), have been used to study theinjurious effects of mechanical ventilation (26, 37, 42). Thesestudies have provided a variety of insights into the damagingeffects of mechanical ventilation by itself as well as in thecontext of preexisting injury (37). Among the variety of path-ological alterations observed in VILI, alterations to the pulmo-nary surfactant system appear to represent one of the mainmechanisms contributing toward lung dysfunction since they1) occur early in the disease process, 2) directly affect lungfunction, and 3) can be prevented by surfactant supplementa-tion (26, 42, 44). Alterations to surfactant also occur in humanpatients with ARDS (13, 14, 40).

Pulmonary surfactant is a complex lipoprotein mixturelining the air spaces of the lung where it helps maintainnormal lung compliance through the reduction of surfacetension at the alveolar surface (12). Specifically, by reduc-ing the surface tension to very low values at low lungvolumes, surfactant prevents alveolar collapse and allowsfor inflation with relatively small changes in pressure. Sur-factant is synthesized within alveolar type II epithelial cells(ATII) where the phospholipids, neutral lipids, and thesurfactant-associated proteins are packaged into lysosomallike-structures, called the lamellar bodies (LBs) (15, 29).Following its packaging into LBs, surfactant is secreted intothe air space by exocytosis at the apical membrane of theATII cell where the LBs quickly unravel and form highlyorganized phospholipid-protein structures capable of reduc-ing surface tension to very low values.

* S. Milos and R. Khazaee contributed equally to this work.Address for reprint requests and other correspondence: R. Veldhuizen,

Lawson Health Research Institute, Rm. F4-110, Western Univ., 268 GrosvenorSt., N6A 4V2, London, ON, Canada (e-mail: rveldhui@uwo.ca).

Am J Physiol Lung Cell Mol Physiol 313: L524–L533, 2017.First published May 25, 2017; doi:10.1152/ajplung.00055.2017.

1040-0605/17 Copyright © 2017 the American Physiological Society http://www.ajplung.orgL524

by 10.220.32.246 on Septem

ber 7, 2017http://ajplung.physiology.org/

Dow

nloaded from

Previous studies on the alterations to surfactant during VILIhave mainly focused on alveolar surfactant (26, 28, 42). Inthese studies, surfactant is obtained via lung lavage after aperiod of high tidal volume ventilation and analyzed for poolsizes, composition, and surface tension-reducing ability. Thesestudies have demonstrated that surfactant obtained from ani-mals with VILI consists of a smaller fraction of functionalsurfactant that is altered in its composition. Most notably,surfactant from animals with VILI contained more cholesterolas compared with surfactant from control animals (26, 31, 42).Furthermore, these changes, as well as the effect of serumproteins that have leaked into the lung during injury, result inan impaired surface tension reducing activity of the recoveredsurfactant (34).

Whereas these previous studies have provided insight intothe alterations of the alveolar surfactant, not much is knownabout the pathological changes to surfactant that occur intra-cellularly. During the injurious ventilation the ATII cell will beexposed to shear stress as well as stretch, both of which mayimpact intracellular pathways associated with LB synthesis andsecretion. It was therefore hypothesized that following VILIthere would be alterations to LB number, composition, bio-physical function, and morphology. This hypothesis was testedusing an established rat model of VILI in which LBs wereisolated from lung tissue for biochemical and functional anal-ysis and, in a separate cohort, visualized in situ using trans-mission electron microscopy (TEM).

MATERIALS AND METHODS

Animals experimentation. A total of 32 male Sprague-Dawley ratsaged 6–10 wk weighing 300–550 g (Charles River, St Constant, PQ,Canada) were used for this study. All procedures were approved bythe Animal Care Committee at the University of Western Ontario andare in agreement with the guidelines of Canadian Council of AnimalCare. Before being used for the experiments, rats were acclimatizedfor a minimum of 3 days in the housing facility, during which timethey had access to standard food and water ad libitum.

Following acclimatization, rats were randomized into one of threeexperimental groups: 1) naïve nonventilated controls (NV), 2) controlventilation (CV), and 3) ventilation-induced lung injury (VILI). Ratsgrouped to CV and VILI were prepared for mechanical ventilation,while NV animals were immediately euthanized. For animals in theCV and VILI groups, animal surgery, intubation, and ventilationprocedures were performed as previously reported (26, 42). Theweight of each rat was recorded, and animals were anesthetized withintraperitoneal injections of ketamine (70–100 mg/kg) and dexme-detomidine (1 mg/kg) in sterile 0.15 M NaCl. The left and rightjugular veins were exposed and catheterized to continuously deliveranesthetic (0.5–2.0 mg·100 g�1·h�1 propofol), neuromuscular blocker(cisatracurium), and sterile fluids (sterile saline with 100 IU heparin/l)at a rate of 0.5–1.0 ml·100 g�1·h�1. The carotid artery was exposedand catheterized to obtain blood samples for blood gas measurements(ABL 500; Radiometer, Copenhagen, Denmark), to monitor heart rateand blood pressure, and to allow continuous delivery of heparinizedsaline using an infusion pump (Model 22; Harvard Apparatus, St.Laurent, PQ, Canada). Following exposure of the trachea, a 14-gaugeendotracheal tube was inserted and tightly secured. Rats were given adose of neuromuscular blocker (100 �g/kg cisatracurium) and wereimmediately connected to a volume cycled rodent ventilator (Model683, Harvard Apparatus) set at a tidal volume (VT) � 8 ml/kg, arespiratory rate (RR) � 54–56 breaths/minute, a positive end expira-tory pressure (PEEP) � 5 cmH2O, and a fraction of inspired oxygen(FIO2

) � 1.0. All animals were given two breath holds (or sighs) to

promote alveolar recruitment and were ventilated at these settings for15 min before sampling of baseline arterial blood gas measurementswith only animals meeting the inclusion criteria of PaO2

� 400 mmHgbeing used for further study.

Animals that met the inclusion criteria were ventilated with eithera CV or an injurious VILI strategy. Rats randomized to CV wereventilated without alteration of the baseline ventilator settings(VT � 8 ml/kg, RR � 54–58 beats/min, PEEP � 5 cmH2O, andFIO2

� 1.0) for 180 min. Rats randomized to the VILI group wereswitched to a second rodent ventilator (Model 683; Harvard Appara-tus) set to deliver ventilation settings of VT � 25–30 ml/kg,RR � 15–18 bets/min, PEEP � 0 cmH2O, and FIO2

� 1.0, for 180min. Peak inspiratory pressure was measured throughout ventilationwith a pressure monitor (Model Criterion 40; Caradyne, Galway,Ireland) connected with the rodent ventilator. Blood gas measure-ments were recorded every 15 to 30 min during the ventilation.Following completion of 180 min of mechanical ventilation, venti-lated rats were euthanized with an intravenous overdose of pentobar-bital sodium (110 mg/kg). Following euthanization, the lungs from therats were either used for LB isolation and biophysical activity (exper-iment 1, n � 6/8 per group) or for electron microscopic analysis ofLBs (experiment 2, n � 3 per group).

Experiment 1: LB isolation and analysis. Isolation of LBs fromwhole lung tissue was performed as previously described (6, 11, 32).All solutions used during LB isolation were 4°C, saline free, andcontained 0.01 M HEPES (pH � 7.4). Briefly, the pulmonary circu-lation was perfused by injection of 10 ml of 0.32 M sucrose solutionfrom the inferior vena cava out a hole cut in the left ventricle of theheart. Following clearance of the pulmonary circulation, lungs wereexcised out of the chest cavity and briefly rinsed in 1.0 M sucrose.Visible large airway tissue was removed from lungs before weighingand tissue homogenization. Subsequently, lungs were cut into smallerpieces and chopped length and width wise using a tissue chopper(McIlwain Tissue Chopper; Mickle Laboratory Engineering). Thetissue was then added to a 1.0 M sucrose solution producing a10–20% (wt/vol) lung homogenate suspension. Homogenization wascarried out using five passes of a serrated, 0.15- to 0.23-mm clearancepestle in a Potter-Elvehjam homogenizer to promote the release ofintracellular contents from lung tissue. The collected homogenate wasfiltered through 4-ply gauze to remove large cellular debris. Smallerdebris was pelleted and removed from the homogenate solution viacentrifugation at 500 g for 15 min at 2°C.

LBs were isolated and purified from homogenate through the use ofdiscontinuous sucrose gradients (32). Briefly, 1.0 M sucrose tissuehomogenate was layered below volumes of 0.32 M sucrose and 0.68M sucrose before centrifugation in a SW28 rotor at 64,000 g for 2 hat 4°C with no brake in a Beckman Couture Optima L-80 XPultra-centrifuge. The LBs containing interfacial band between 0.32and 0.68 M sucrose was collected. With the aid of a refractometer(ABBE Refractometer, model A300A; Fisher Scientific, Waltham,MA), concentrated 1.0 M sucrose was added to the collected interfa-cial materials to produce a sucrose solution of 0.58 M. The LB-containing 0.58 M sucrose solution was then layered beneath 0.32 Msucrose and 0.45 M sucrose before being centrifuged again at identicalsettings. Purified LBs were collected at the 0.45–0.58 M sucroseinterfacial band. Subsequently, the isolated LB solution was diluted to0.2 M sucrose and LBs were pelleted by centrifuging the collectedmaterials at 40,000 g for 15 min at 4°C. The resulting LB pellet wasresuspended in 300 �l of saline and frozen at �20°C for subsequentanalysis.

Three separate animals were used to confirm the LB isolation andpurity by TEM. Briefly, the isolated LBs were collected in 0.2 Msucrose containing 0.1 M sodium cacodylate buffer (pH 7.3) and 1.2%glutaraldehyde and 1.2% formaldehyde as primary fixatives (33).LB-containing samples were centrifuged at 40,000 g for 15 min at 4°Cto produce a pellet. Further processing was carried out as previouslydescribed (9) including postfixation for 1 h with 1% osmium tetroxide

L525LAMELLAR BODIES IN VENTILATION-INDUCED LUNG INJURY

AJP-Lung Cell Mol Physiol • doi:10.1152/ajplung.00055.2017 • www.ajplung.org

by 10.220.32.246 on Septem

ber 7, 2017http://ajplung.physiology.org/

Dow

nloaded from

in 0.1 M Cacodylate buffer and en bloc staining overnight (12–18 h)at 4–8 C with half-saturated aqueous uranyl acetate. Specimens weredehydrated in an ascending series of acetone solutions and embeddedin Epon-Araldite resin at 60°C for 3–5 days. Ultra-thin (70 nm)sections were cut (Ultramicrotome Reichert-Jung Ultracut E; LeicaMicrosystems, Wetzlar, Denmark) and poststained with 2% uranylacetate and lead citrate. TEM imaging of LB samples was performedat low and high magnifications (Fig. 1). Low-magnification micro-graphs showed isolated LB organelles without the presence of otheridentifiable cellular ultrastructures, such as mitochondria, or the pres-ence of other surfactant components, such as tubular myelin. Highmagnifications of isolated LBs displayed characteristic features suchas dense osmiophilic folds of phospholipids surrounded by an intactlimiting membrane.

For LB isolations used for analysis, phospholipid pools weremeasured after lipid extraction using a modified Duck-Chong phos-pholipid-phosphorus assay (4, 8). The amount of surfactant recoveredin the isolated LB fractions was expressed as milligrams phospholipidper unit body weight. Total cholesterol levels in isolated LBs werequantified using an enzymatic colorimetric Amplex Red cholesterolassay kit, as per manufacturer’s instructions (Thermo Fisher Scien-tific, Waltham, MA).

The biophysical function of the isolated LBs was assessed using aconstrained sessile drop surfactometer (CSD) as previously described(28, 38). Briefly, aliquots of LBs samples were centrifuged at 21,000g for 15 min at 2°C to concentrate samples at 2 mg/ml phospholipidin buffer (140 mM NaCl, 2.5 mM HEPES, and 1.5 mM CaCl2, pH7.4). Glass beads were added to samples to promote mixing, andsamples were heated at 37° C for at least 1 h before biophysicalassessment on the CSD. A 9-�l drop of sample was placed on theCSD pedestal within an environmental controlled chamber at 37°Cand atmospheric humidity. Samples were allowed to equilibrate over3 minutes without manipulation of the sample drop. Samples werethen exposed to 20 repeated compression-expansion cycles on theCSD at a rate of 30 cycles/min. Drop images were recorded duringcompression/expansion cycles and were analyzed by AxisymmetricDrop Shape Analysis software to obtain an accurate measurement ofsurface tension (19). LB samples from each rat had three to fourtechnical replicates exposed to dynamic compression-expansion cy-cles.

Experiment 2: TEM analysis. In the second experiment, nine ratswere used to assess the impact of various ventilation strategies on LBnumber and area within the ATII cell as well as LB morphology insitu within the lung tissue. Three rats from each ventilation strategyhad lungs processed as previously described (10). Briefly, at the endof ventilation, lungs were fixed in situ at a pressure of 20 cm watercolumn by intratracheal instillation using the primary fixative solu-tion. The fixative solution contained 1.5% paraformaldehyde and1.5% glutaraldehyde in 0.1 M cacodylate buffer at 4°C (pH 7.3) (33).Following filling the airways by instillation of the primary fixativesolution via the trachea into the lungs, the trachea was clamped, andthe intact lungs were submerged in the same fixation solution for 24h (at 4°C). Further processing and sampling of lungs was carried outas described previously (10). Following the separation into the rightand left single lung, each lung was cut from the apex to the base intoparallel slices with a thickness of 3 mm using a tissue slicer. Randomtissue sampling was done using a transparent Plexiglas grid with5-mm holes placed over the collected slices. Parts of the tissueunderlying the holes were excised using a 5-mm colonic biopsy punchand cut into small 1- to 2-mm blocks.

Further tissue processing was carried out as described previously(9). Blocks were postfixed for 2 h in 1% osmium tetroxide in 0.1 Msodium cacodylate buffer (pH 7.3) and then stained en bloc with halfsaturated uranyl acetate for 12–18 h at 4°C. Samples were dehydratedin an ascending acetone series and embedded in Epon-Araldite resinat 60°C for 3–5 days (22). Ultra-thin (70 nm) sections were cut(Ultramicrotome Reichert-Jung Ultracut E; Leica Microsystems) andpoststained with 2% uranyl acetate and 2% lead citrate.

Imaging of isolated LBs and ATII epithelial cells was carried outusing a Transmission Electron Microscope CM10 (Philips ElectronOptics, Eindhoven, The Netherlands). Morphological and quantitativeimage analysis in all experimental groups was carried out via Image-Pro Premier software (Media Cybernetics, Rockville, MD) to obtainthree types of measurements: 1) relative area and number of LBswithin the ATII cells; 2) morphological features of LBs includingaspect ratio, roundness, mean diameter, and 3) intensity of intra-LB-stained materials as a representation of their phospholipid content. Intotal, 60 images/experimental group for morphological analysis and90 images per experimental group for quantitative analysis wereanalyzed. For distribution histograms, LB area and staining intensityvalues were separated into seven intervals and the percentage ofimages in each interval was expressed for each experimental group.

Statistical analysis. Data are expressed as means � SE. All statis-tics were performed using statistical analysis software GraphPadPrism (GraphPad Software, La Jolla, CA). Comparisons betweenventilation strategies were made using a one-way ANOVA with aTukey’s post hoc test. P � 0.05 was considered statistically signifi-cant.

RESULTS

Physiology. For physiological data, animals used for thebiophysical measurements and electron microscopy analysiswere combined for each of the three experimental groups.Body weight was not significantly different among the groups(data not shown). The arterial oxygenation values over the180-min ventilation period of the two mechanically ventilatedgroups are shown in Fig. 2. There were no significant differ-ences in arterial oxygenation between CV and VILI at any timefrom baseline to 135 min. At 150 min through to the end of theduration of mechanical ventilation, animals exposed to VILIhad significantly reduced arterial oxygenation compared withCV rats. Rats exposed to CV did not experience any significantdecrease in oxygenation at any time point as compared withbaseline, while rats exposed to VILI had a significant decline in

Fig. 1. Representative transmission electron microscopy (TEM) images ofisolated lamellar bodies (LBs) from nonventilated (NV) rat lungs: image showsintact LBs at magnifications of �4,600 (A) and �92,000 (B) without otheridentifiable structures present. Scale bar in B � 100 nm.

L526 LAMELLAR BODIES IN VENTILATION-INDUCED LUNG INJURY

AJP-Lung Cell Mol Physiol • doi:10.1152/ajplung.00055.2017 • www.ajplung.org

by 10.220.32.246 on Septem

ber 7, 2017http://ajplung.physiology.org/

Dow

nloaded from

oxygenation as compared with baseline at times 150 through-out to the end of ventilation.

Table 1 displays additional physiological parameters as-sessed during ventilation. At the baseline, there was no differ-ence in any measured physiological parameter between the ratsexposed to either ventilation strategy. Following 180 min ofventilation, Peak inspiratory pressure was significantly in-creased in VILI-exposed animals compared with CV. Further-more, following 180 min of ventilation, blood pressure wasreduced, and peak inspiratory pressure was increased as com-pared with baseline in all ventilated rats, while heart rate wasonly increased in CV rats as compared with baseline.

Experiment 1: LB isolation and analysis. Following isola-tion, the isolated LB samples were analyzed for the quantity oftheir phospholipid contents and the relative cholesterol content(%phospholipid). As shown in Fig. 3A, CV rats ventilated for180 min had significantly more total phospholipid levels re-covered in the LB pellet following the isolation process ascompared with both NV rats and VILI rats. However, there wasno difference in recovered LB phospholipid content betweenspontaneously breathing NV animals and those exposed toVILI. The percentage of cholesterol, relative to phospholipid,was not different among the three groups (Fig. 3B).

Subsequently, the isolated LB samples had their biophysicalproperties assessed on the CSD. Representative surface ten-sions vs. surface area isotherms of isolated LBs are shown inFig. 4. LBs isolated from NV and CV rats displayed lowsurface tensions upon the initial compression (�5 mN/m),while rats exposed to VILI had an impaired ability to reducesurface tension, reaching a surface tension of 15 mN/m at cycle1 (Fig. 4). Repeated dynamic surface area cycling lowered theminimum surface tension achieved by LB samples isolated

from all ventilation groups compared with the first respectivedynamic compression cycle (cycle 1 vs. cycle 10). In addition,isotherms constructed at different compression cycles demon-strated that LBs isolated from rats grouped to NV generally hada substantial reduction in hysteresis area following repeatedcycling, a phenomenon not observed in LBs isolated frommechanically ventilated rats. Whereas these representative iso-therms demonstrate the overall surfactant surface tensionchanges achieved during cyclical compression and expansionof the representative samples, the quantification and statisticalcomparisons of minimum surface tension and other biophysicalparameters from all LB samples are shown in Fig. 5 and Table2, respectively.

Minimum surface tensions of LB samples obtained from ratsfrom the three experimental groups are shown in Fig. 5. LBs

Fig. 2. Arterial oxygenation (mmHg) of rats exposed to 3 h of controlventilation (CV) and ventilation-induced lung injury (VILI). Data expressed asmeans � SE. *P � 0.01 vs. CV; P � 0.001 vs. baseline time; n �9/treatment group.

Table 1. Lung physiology of rats at baseline and after exposure to 3 h of mechanical ventilation

Blood Analysis Control Ventilation VILI

Time point, min Baseline 180 min Baseline 180 min

PIP, cmH2O 11.11 � 0.45 15.33 � 0.2† 10.89 � 0.46 31.61 � 1.18*,†PaCO2

, mmHg 34.8 � 1.6 37.3 � 1.3 32.2 � 2.1 44.4 � 3.5

Blood pressure, cmH2O 165.8 � 8.6 80.2 � 3.8† 159.2 � 6.7 48.0 � 5.1*,†Heart rate, beats/min 259.7 � 4.6 270.8 � 6.6† 247.9 � 12.0 243.1 � 15.5Blood pH 7.40 � 0.01 7.33 � 0.02 7.41 � 0.02 7.28 � 0.02

Values are means � SE; n � 9/group. VILI, ventilation-induced lung injury; PIP, peak inspiratory pressure. *P � 0.01 vs. control ventilation (CV). †P � 0.05vs. baseline time.

Fig. 3. Phospholipid (mg/kg body wt; A) and cholesterol as a percentage ofphospholipids (B) recovered in isolated lamellar bodies from NV rats, or ratsexposed to 3 h of CV or VILI. #P � 0.05 vs. NV; *P � 0.05 vs. CV; n �6-8/treatment group.

L527LAMELLAR BODIES IN VENTILATION-INDUCED LUNG INJURY

AJP-Lung Cell Mol Physiol • doi:10.1152/ajplung.00055.2017 • www.ajplung.org

by 10.220.32.246 on Septem

ber 7, 2017http://ajplung.physiology.org/

Dow

nloaded from

from rats exposed to VILI had significantly higher minimumsurface tensions achieved compared with associated controls.The impairment associated with VILI LB samples was statis-tically different at cycles 1–10 and cycles 1–14 and 18 com-pared with the minimum surface tension of surfactant obtained

from CV and NV rats, respectively (Fig. 5). There were nosignificant differences in minimum surface tension achievedupon compression of isolated LBs between control NV and CVsamples. Comparing minimum surface tensions achieved dur-ing repeated compression expansions, it was demonstrated thatthere were significant reductions in minimum surface tensionachieved over the course of dynamic cycling as compared withcycle 1 for both NV (cycle 2 and onward)- and VILI (cycle 3and onwards)-exposed animals.

The additional outcomes of the biophysical assessment ofthe isolated LBs are shown in Table 2. First, there were nosignificant differences between VILI and controls with respectto the surface tension measured following 120- to 180-s LB

Fig. 4. Representative surface tension(mN/m) vs. surface area (fraction of the initialarea) isotherms at dynamic compression-expansion cycles 1 and 10 of lamellar bodiesisolated from NV rats (top) or rats exposed to 3h of CV (middle) or VILI (bottom). Withineach isotherm, open symbols represent com-pression of the surface film and closed symbolsrepresent the expansion of the film.

Fig. 5. Minimum surface tension (mN/m) of isolated lamellar bodies during 20dynamic compression-expansion cycles from NV rats, or rats exposed to 3 h ofCV or VILI. #P � 0.05: VILI vs. NV; *P � 0.05: VILI vs. CV; n �6-8/treatment group.

Table 2. Maximal area compression, adsorption, andmaximum surface tension of lamellar bodies isolated fromrats exposed to ventilation

NV CV VILI

Surface tension followingadsorption, mN/m 23.2 � 0.3 22.9 � 0.2 24.1 � 0.4

Maximum compression, %originalarea 73.6 � 0.6 71.7 � 0.8 73.0 � 0.7

Maximum surface tension, mN/mCycle 1 34.0 � 0.8 32.4 � 0.6 32.3 � 0.6Cycle 10 34.3 � 0.4 33.6 � 0.6 33.8 � 1.0

Values are means � SE; n � 6–8/treatment group.

L528 LAMELLAR BODIES IN VENTILATION-INDUCED LUNG INJURY

AJP-Lung Cell Mol Physiol • doi:10.1152/ajplung.00055.2017 • www.ajplung.org

by 10.220.32.246 on Septem

ber 7, 2017http://ajplung.physiology.org/

Dow

nloaded from

sample adsorption on the CSD pedestal. Second, all LB sam-ples regardless of origins were compressed to similar extentsduring the functional assessment, as indicated by maximumcompression at cycle 1. Lastly, there was no difference in themaximum surface tensions achieved between LB samples iso-lated from rats exposed to various ventilation strategies ateither cycle. Furthermore, there was no difference in maximumsurface tension achieved comparing cycle 1 and 10 within eachventilation strategy.

Experiment 2: in situ imaging of LBs. TEM imaging ofATIIs and LBs was performed on lung tissue samples from allthree experimental groups. Representative images for allgroups are shown in Fig. 6. Relatively low magnificationsmicrographs provided images of LBs within the ATII struc-tures. Higher magnifications were used to visualize the variousmorphological details of LBs (Fig. 6B).

Quantitation of the LB number, size, and area occupiedwithin the ATII cells is shown in Fig. 7. Statistical comparisonsrevealed that the number of LBs in ATII cells significantlyincreased in CV compared with NV group. The number of LBsin VILI animals decreased compared with both NV and CVgroups by approximately threefold (Fig. 7A). The relative areaof ATII occupied by the LBs (Fig. 7B) was significantly lowerin animals from the VILI group as compared with both CV andNV groups. In addition, the area occupied in the CV showed asignificant increase compared with the NV group.

Further morphological comparisons of LBs of rats exposedto various ventilation strategies are shown in Table 3. Com-paring the size and shape aspects of LBs visualized from all

three ventilation strategies, it was found there were no signif-icant differences among groups regarding roundness and aspectratio, while significant differences were observed in mean LBarea and diameter between CV and NV groups. Stainingintensity reflects the mean values of intensity levels per pixelon the images of LBs in each group. Black color, whichrepresents a fully stained area, has a value of zero and white(absence of any staining) represents 255 in this measurement.The significant increase in the gray scale in VILI groupindicates a decrease of staining compared with the LBs fromCV group (Table 3). To further explore the significant changesin morphology, distribution histograms of LB area and stainingintensity are shown in Fig. 8. Increases in LB area overall inthe CV group are shown to be due to a decrease in therelatively small LBs and an increase in larger (1.0 �m andlarger) LBs (Fig. 8A). Differences in staining intensity are dueto an overall increase in staining values (i.e., lighter staining) inthe LBs from the NV group as compared with the two othergroups (Fig. 8B).

DISCUSSION

Mechanical ventilation is a common supportive therapywith, under some conditions, the potential of damaging thelung (2). This injury, termed VILI in animals, is in partmediated through alterations to the pulmonary surfactant sys-tem (26, 28, 41, 42). A variety of studies have examined theimpact of VILI on surfactant’s pool sizes and biophysicalfunction by examining the extracellular surfactant obtained via

Fig. 6. Representative electron micrographs from each of the experimental groups at two different magnifications. Top row: alveolar type II epithelial cells (ATII)with LBs inside, �5,800; scale bar � 2 um. Bottom row: LB, �64,000; scale bar � 100 nm.

L529LAMELLAR BODIES IN VENTILATION-INDUCED LUNG INJURY

AJP-Lung Cell Mol Physiol • doi:10.1152/ajplung.00055.2017 • www.ajplung.org

by 10.220.32.246 on Septem

ber 7, 2017http://ajplung.physiology.org/

Dow

nloaded from

lung lavage. The current study has added to these previousfindings by examining the effect of VILI on the intracellularsource of surfactant, the LB. It was hypothesized that therewould be alterations to LB number, basic composition, bio-physical function, and morphology following exposure toVILI. The results demonstrated that minimum surface tensionreduction of unraveled LBs isolated from rats exposed to VILIwas significantly impaired during early dynamic compression-expansion cycles compared with noninjured ventilator con-trols. Furthermore, LB number and area occupied within theATII decrease after VILI compare with CV and NV groups.Taken together, it is concluded that VILI causes significant

alterations to LBs and we speculate that these observedchanges contribute to impairments in surface tension reductionwithin the alveolar hypophase.

To investigate the impact of VILI on intracellular surfactantwithin ATII, we used an established model of VILI (26, 42).Lung injury in this model was verified by indicators of pul-monary function, such as reduced PaO2

:FIO2and other physio-

logical parameters similar to those previous studies on VILI.Subsequently, two aspects of the LBs were investigated; first,we examined the biophysical function of isolated LBs using theconstrained sessile drop surfactometer, and second, we per-formed a morphological analysis of LBs through TEM of fixedlung tissue. Together, these complementary approaches al-lowed for a thorough examination of the LB within the differ-ent experimental groups and support our overall conclusionthat some of the alterations of surfactant due to mechanicalventilation are initiated intracellularly.

One of the novel aspects of our study is that analysis of LBbiophysical function on the CSD following VILI. To ourknowledge, this analysis represents one of the first detailedbiophysical analyses of LB function including adsorption andsurface tension reduction during dynamic compression andexpansion cycling. Previous studies, such as those by Haller’s

Fig. 7. Analysis of LB number/ATII micrograph (A) %area within the ATII(B). Values are expressed as mean � SE. #P � 0.05 vs. NV; *P � 0.05 vs. CV;n � 3/group.

Table 3. Morphological analysis of LBs visualized withinATII from the three ventilation groups

Lamellar body morphologicalparameter NV CV VILI

Average intensity per pixel† 110.3 � 6.7 97.7 � 13.8 125.3 � 5.4*Area, �m2 0.37 � 0.01 0.81 � 0.14# 0.54 � 0.12Diameter, �m 0.64 � 0.01 0.92 � 0.08# 0.75 � 0.1Circularity‡ 1.12 � 0.01 1.12 � 0.02 1.40 � 0.2Aspect ratio§ 1.40 � 0.03 1.37 � 0.04 1.39 � 0.06

Values are means � SE; n � 3/group. ATII, alveolar type II epithelial cells.*P � 0.05 vs. CV, #P � 0.05 vs. NV. †Measure of brightness of each pixelon LB image. ‡Circularity is calculated as perimeter^2/(4 � pi � area).§Aspect ratio � ratio or major axis to minor axis of an ellipse.

Fig. 8. Analysis of LB area (A) and %LB intensity distribution (B). Individualimages of LBs measured and the percentage in each area (A) or intensity (B)interval were calculated. The intensity is measured as average intensity perpixel in which a black color, represents a fully stained area with a value of zeroand white (absence of any staining) represents 255. Values are expressed asmeans � SE; total number of images analyzed was 60–90/group. *P � 0.05vs. NV; #P � 0.05 vs. CV; n � 3/group.

L530 LAMELLAR BODIES IN VENTILATION-INDUCED LUNG INJURY

AJP-Lung Cell Mol Physiol • doi:10.1152/ajplung.00055.2017 • www.ajplung.org

by 10.220.32.246 on Septem

ber 7, 2017http://ajplung.physiology.org/

Dow

nloaded from

group, have elegantly shown the adsorptive capacity and un-raveling at the air-liquid interface of LBs (17, 18); our studybuilds on these findings by subjecting LBs isolated in vivo todynamic compression-expansion cycles. It demonstrates thatisolated LBs, under conditions which have promoted its un-ravelling, can reduce surface tension values to near zero valuessimilar to isolated surfactant from the alveolar space. Further-more, our experiment examined this biophysical function fol-lowing exposure to VILI and demonstrated impairment in thissurface tension activity. Several studies have demonstratedcompositional and structural differences in the LBs followinginduction of lung injury (21, 22); however, our study is the firstto do so from a biophysical perspective.

In our experiments, impairments in surface tension reductionwere observed in VILI samples during the first few compres-sion-expansion cycles; however, continued cycling improvedthese minimum surface tensions achieved. Thus LBs isolatedfrom VILI-exposed rats could overcome their initial impair-ment in surface tension reduction by repeated compressionsand expansions, as would occur during continuous respiration.The phenomenon of improved surfactant surface tension re-duction during repeated lateral compression is likely explainedby interfacial film remodeling, involving the removal (or“squeeze out”) of components interfering with reaching lowsurface tension (20). The process, includes the removal offluidizing, unsaturated phospholipid, cholesterol, and poten-tially other components from the surface film resulting in theformation of a dipalmitoylphosphatidylcholine-enriched filmcapable of reaching low surface tensions (20). We previouslyreported that elevated cholesterol was a contributing factor tothe inhibition of alveolar surfactant obtained by lavage byaffecting the fluidity of the surfactant film (31, 42). In thecurrent study, analysis of the relative cholesterol levels in theisolated LB fraction demonstrated no differences amonggroups. Our data lead us to suggest that upon film formation atthe air-liquid interface, LB surfactant isolated from NV or CVrats could immediately form a dipalmitoylphosphatidylcho-line-enriched film upon the first compression, while LBs fromVILI-exposed rats required several compressions to result in amore purified surface film. However, this functional impair-ment is not due to elevated levels of cholesterol.

A second novel aspect of our investigation was assessing theintra-alveolar LB surfactant content through TEM image anal-ysis in VILI. The results obtained included reduced numbersand relative area occupied by LBs in VILI, as well as structuralassessments such as shape and staining intensity. Distributionhistograms of the LBs showed that the VILI group had rela-tively more smaller LBs and that these LBs were lighter incolor after staining, suggesting lower phospholipid content.The previous investigations into LBs in the context of lunginjuries have mostly focused on lung transplantation and isch-emia reperfusion (7, 10, 23). These studies, using extensivestereological analyses, showed an association of decreasedlevels of LBs within the type II cell with lung injury. In bothscenarios, it is feasible that the lower number of LBs is theresult extensive exocytosis of surfactant phospholipids andproteins. Support for this notion stems from the numerousstudies investigating surfactant secretion, in which alveolarstretch appears to be one of the most effective signals forexocytosis (1, 27). Furthermore, in a previous study from ourgroup, we showed an increase in alveolar pool size of surfac-

tant following VILI, which was followed by increased alveolarmetabolism and inactivation of the surfactant (26). Togetherwith the current observations of a decreased staining of LBs inVILI and impaired function of the isolated LBs, the resultssupport the conclusion that the secretion of surfactant due tomechanical ventilation cannot be compensated for by synthe-sis/secretion of de novo LBs in these injurious conditions.

The focus of our analysis of the LBs was on the biophysicalfunction and morphology, with only a limited compositionalanalysis of our samples. Nevertheless, both the functional andmorphological alterations observed in the LBs from the VILIgroup are, by definition, due to compositional alterations andmodifications. The literature on compositional analysis of LBs,or on proteins and genes involved in surfactant and LB bio-synthesis, in the context of VILI is relatively scarce. Forexample, gene array studies have identified many genes thatare altered in VILI; however, the genes identified are repre-sentative of all a mixture of all lung cells, in which changes toLB metabolism may not provide a strong enough signal to bedetected (45). Indeed, the majority of genes identified withthese techniques involve generic pathways of inflammation,wound healing, stress response, among others (45). A moretargeted approach has yielded a bit more insight and should beused in future studies. For example, we have previously re-ported decreases of mRNA levels of surfactant proteins B andC in an isolated rat model of VILI (39), which could have animpact on the biophysical function. In addition, stretching of atype II cell-like cell line has yielded information on phospho-lipase A2 activation (25), which plays a crucial role in surfac-tant metabolism, as well as the induction of synthesis of themain phospholipid of surfactant, phosphatidylcholine (30). Interms of LB formation and morphological alterations, the roleof surfactant lipid transporter ABCA3 should be investigatedfurther. A conditional null mutation of this transporter resultedin altered LB formation, morphological alterations, and alteredPL composition (3). Although these changes did not com-pletely replicate our VILI observations, they do support a rolefor ABCA3 in lung injury. Furthermore, it has been shown thatheterozygotes for the null mutation of this protein were moresusceptible to VILI (16). Overall, the exact genes, proteins, andpathways involved in mediating the functional and structuralalteration of LBs in VILI are unknown. It is speculated thatalterations in the production of phospholipids and surfactantproteins, as well as changes in ABCA3 lipid transporter, likelycontribute to the LB morphological and biophysical alterationsobserved between VILI-exposed samples and controls.

It should be noted that there are inherent limitations associ-ated with our studies. For example, our model of VILI is basedon a short time period of high tidal volume ventilation to reflecta clinical condition that progresses with lower tidal volumesover prolong periods of time in human patients who have anunderlying lung disease. Nevertheless, several previous obser-vations in animal models of VILI, such as ventilation-inducedlung inflammation (biotrauma) and elevated cholesterol withinsurfactant, have subsequently been confirmed in clinical situ-ations (31, 36, 46). Our analysis was limited to a time point atwhich lung injury was severe; it would be interesting toanalyze LBs at a stage before significant lung injury has beenreached to assess if the changes in LBs precede those physio-logical changes. Such studies are considered a future direction.In addition, our visualization of LBs and ATII were obtained

L531LAMELLAR BODIES IN VENTILATION-INDUCED LUNG INJURY

AJP-Lung Cell Mol Physiol • doi:10.1152/ajplung.00055.2017 • www.ajplung.org

by 10.220.32.246 on Septem

ber 7, 2017http://ajplung.physiology.org/

Dow

nloaded from

from two-dimensional image analysis; future studies shouldconsider complementary volumetric and topographic stereol-ogy of LBs and ATII.

In conclusion, whereas previous studies have primarily fo-cused on the role of alveolar surfactant in VILI, this study hasexpanded this knowledge by investigating the intracellularimpairments of surfactant. Our studies showed a decrease inLB content within the lungs of animals with VILI and thatthese remaining LBs had impaired biophysical properties whenanalyzed on a constrained sessile drop surfactometer. Thesealterations suggest that overstretching and collapse due toinjurious ventilation promote secretion of LBs but do not resultin increased synthesis of LBs, and as a consequence, subse-quent impairment of the secreted surfactant during ventilationcannot be compensated or overcome by further secretion of denovo surfactant.

ACKNOWLEDGMENTS

We thank Drs. Fred Possmayer and Jim Lewis for helpful discussions.

GRANTS

Funds for these studies were provided by an operating grant from theCanadian Institutes of Health Research and Grants-in-Aid from the OntarioThoracic Society.

DISCLOSURES

No conflicts of interest, financial or otherwise are declared by the authors.

AUTHOR CONTRIBUTIONS

S.M., R.K., R.B.G., Y.Y.Z., C.M.Y., and R.A.V. conceived and designedresearch; S.M., R.K., L.A.M., and K.N. performed experiments; S.M., R.K.,L.A.M., K.N., and R.A.V. analyzed data; S.M., R.K., L.A.M., K.N., R.B.G.,Y.Y.Z., C.M.Y., and R.A.V. interpreted results of experiments; S.M. and R.K.prepared figures; S.M. and R.K. drafted manuscript; S.M., R.K., L.A.M., K.N.,R.B.G., Y.Y.Z., C.M.Y., and R.A.V. approved final version of manuscript;K.N., R.B.G., Y.Y.Z., C.M.Y., and R.A.V. edited and revised manuscript.

REFERENCES

1. Andreeva AV, Kutuzov MA, Voyno-Yasenetskaya TA. Regulation ofsurfactant secretion in alveolar type II cells. Am J Physiol Lung Cell MolPhysiol 293: L259–L271, 2007. doi:10.1152/ajplung.00112.2007.

2. Antonelli M, Azoulay E, Bonten M, Chastre J, Citerio G, Conti G, DeBacker D, Lemaire F, Gerlach H, Hedenstierna G, Joannidis M,Macrae D, Mancebo J, Maggiore SM, Mebazaa A, Preiser JC, PuginJ, Wernerman J, Zhang H. Year in review in Intensive Care Medicine2009. Part III: mechanical ventilation, acute lung injury and respiratorydistress syndrome, pediatrics, ethics, and miscellanea. Intensive Care Med36: 567–584, 2010. doi:10.1007/s00134-010-1781-0.

3. Besnard V, Matsuzaki Y, Clark J, Xu Y, Wert SE, Ikegami M,Stahlman MT, Weaver TE, Hunt AN, Postle AD, Whitsett JA. Con-ditional deletion of Abca3 in alveolar type II cells alters surfactanthomeostasis in newborn and adult mice. Am J Physiol Lung Cell MolPhysiol 298: L646–L659, 2010. doi:10.1152/ajplung.00409.2009.

4. Bligh EG, Dyer WJ. A rapid method of total lipid extraction andpurification. Can J Biochem Physiol 37: 911–917, 1959. doi:10.1139/o59-099.

5. Brower RG, Matthay MA, Morris A, Schoenfeld D, Thompson BT,Wheeler A; Acute Respiratory Distress Syndrome Network. Ventila-tion with lower tidal volumes as compared with traditional tidal volumesfor acute lung injury and the acute respiratory distress syndrome. N EnglJ Med 342: 1301–1308, 2000. doi:10.1056/NEJM200005043421801.

6. Chander A, Johnson RG, Reicherter J, Fisher AB. Lung lamellarbodies maintain an acidic internal pH. J Biol Chem 261: 6126–6131, 1986.

7. Dreyer N, Mühlfeld C, Fehrenbach A, Pech T, von Berg S, Nagib R,Richter J, Wittwer T, Wahlers T, Ochs M. Exogenous surfactantapplication in a rat lung ischemia reperfusion injury model: effects onedema formation and alveolar type II cells. Respir Res 9: 5, 2008.doi:10.1186/1465-9921-9-5.

8. Duck-Chong CG. A rapid sensitive method for determining phospholipidphosphorus involving digestion with magnesium nitrate. Lipids 14: 492–497, 1979. doi:10.1007/BF02533467.

9. Fehrenbach H. Studying lung ultrastructure. In: Methods in PulmonaryResearch (Uhlig S, Taylor AE, eds.). Basel, Switzerland: Birkhäuser,1998, p. 429–454. doi:10.1007/978-3-0348-8855-4_17.

10. Fehrenbach H, Wahlers T, Ochs M, Brasch F, Schmiedl A, Hirt SW,Haverich A, Richter J. Ultrastructural pathology of the alveolar type IIpneumocytes of human donor lungs. Electron microscopy, stereology,and microanalysis. Virchows Arch 432: 229–239, 1998. doi:10.1007/s004280050160.

11. Gil J, Reiss OK. Isolation and characterization of lamellar bodies andtubular myelin from rat lung homogenates. J Cell Biol 58: 152–171, 1973.doi:10.1083/jcb.58.1.152.

12. Goerke J. Pulmonary surfactant: functions and molecular composi-tion. Biochim Biophys Acta 1408: 79–89, 1998. doi:10.1016/S0925-4439(98)00060-X.

13. Gregory TJ, Longmore WJ, Moxley MA, Whitsett JA, Reed CR,Fowler AA III, Hudson LD, Maunder RJ, Crim C, Hyers TM.Surfactant chemical composition and biophysical activity in acute respi-ratory distress syndrome. J Clin Invest 88: 1976–1981, 1991. doi:10.1172/JCI115523.

14. Günther A, Ruppert C, Schmidt R, Markart P, Grimminger F,Walmrath D, Seeger W. Surfactant alteration and replacement in acuterespiratory distress syndrome. Respir Res 2: 353–364, 2001. doi:10.1186/rr86.

15. Haagsman HP, van Golde LM. Synthesis and assembly of lung surfac-tant. Annu Rev Physiol 53: 441–464, 1991. doi:10.1146/annurev.ph.53.030191.002301.

16. Herber-Jonat S, Mittal R, Huppmann M, Hammel M, Liebisch G,Yildirim AÖ, Eickelberg O, Schmitz G, Hrabé de Angelis M, FlemmerAW, Holzinger A. Abca3 haploinsufficiency is a risk factor for lunginjury induced by hyperoxia or mechanical ventilation in a murine model.Pediatr Res 74: 384–392, 2013. doi:10.1038/pr.2013.127.

17. Hobi N, Giolai M, Olmeda B, Miklavc P, Felder E, Walther P, Dietl P,Frick M, Pérez-Gil J, Haller T. A small key unlocks a heavy door: theessential function of the small hydrophobic proteins SP-B and SP-C totrigger adsorption of pulmonary surfactant lamellar bodies. Biochim Bio-phys Acta 1863: 2124–2134, 2016. doi:10.1016/j.bbamcr.2016.04.028.

18. Hobi N, Siber G, Bouzas V, Ravasio A, Pérez-Gil J, Haller T.Physiological variables affecting surface film formation by native lamellarbody-like pulmonary surfactant particles. Biochim Biophys Acta 1838:1842–1850, 2014. doi:10.1016/j.bbamem.2014.02.015.

19. Kalantarian A, Saad SM, Neumann AW. Accuracy of surface tensionmeasurement from drop shapes: the role of image analysis. Adv ColloidInterface Sci 199-200: 15–22, 2013. doi:10.1016/j.cis.2013.07.004.

20. Keating E, Zuo YY, Tadayyon SM, Petersen NO, Possmayer F,Veldhuizen RA. A modified squeeze-out mechanism for generating highsurface pressures with pulmonary surfactant. Biochim Biophys Acta 1818:1225–1234, 2012. doi:10.1016/j.bbamem.2011.12.007.

21. Kirkland JB, Bray TM. The effect of 3-methylindole on the quantity andfunctional quality of lung surfactant. Can J Physiol Pharmacol 66:895–900, 1988. doi:10.1139/y88-146.

22. Knudsen L, Boxler L, Mühlfeld C, Schaefer IM, Becker L, BussingerC, von Stietencron I, Madershahian N, Richter J, Wahlers T, WittwerT, Ochs M. Lung preservation in experimental ischemia/reperfusioninjury and lung transplantation: a comparison of natural and syntheticsurfactants. J Heart Lung Transplant 31: 85–93, 2012. doi:10.1016/j.healun.2011.10.002.

23. Knudsen L, Waizy H, Fehrenbach H, Richter J, Wahlers T, WittwerT, Ochs M. Ultrastructural changes of the intracellular surfactant pool ina rat model of lung transplantation-related events. Respir Res 12: 79, 2011.doi:10.1186/1465-9921-12-79.

24. Lee WL, Slutsky AS. Ventilator-induced lung injury and recommenda-tions for mechanical ventilation of patients with ARDS. Semin Respir CritCare Med 22: 269–280, 2001. doi:10.1055/s-2001-15784.

25. Letsiou E, Kitsiouli E, Nakos G, Lekka ME. Mild stretch activatescPLA2 in alveolar type II epithelial cells independently through theMEK/ERK and PI3K pathways. Biochim Biophys Acta 1811: 370–376,2011. doi:10.1016/j.bbalip.2010.12.007.

26. Maruscak AA, Vockeroth DW, Girardi B, Sheikh T, Possmayer F,Lewis JF, Veldhuizen RA. Alterations to surfactant precede physiologicaldeterioration during high tidal volume ventilation. Am J Physiol Lung CellMol Physiol 294: L974–L983, 2008. doi:10.1152/ajplung.00528.2007.

L532 LAMELLAR BODIES IN VENTILATION-INDUCED LUNG INJURY

AJP-Lung Cell Mol Physiol • doi:10.1152/ajplung.00055.2017 • www.ajplung.org

by 10.220.32.246 on Septem

ber 7, 2017http://ajplung.physiology.org/

Dow

nloaded from

27. Mason RJ, Voelker DR. Regulatory mechanisms of surfactant secre-tion. Biochim Biophys Acta 1408: 226–240, 1998. doi:10.1016/S0925-4439(98)00070-2.

28. Milos S, Qua Hiansen J, Banaschewski B, Zuo YY, Yao LJ, McCaigLA, Lewis J, Yamashita CM, Veldhuizen RA. The effect of diet-induced serum hypercholesterolemia on the surfactant system and thedevelopment of lung injury. Biochem Biophys Reports 7: 180–187, 2016.doi:10.1016/j.bbrep.2016.06.009.

29. Oosterlaken-Dijksterhuis MA, van Eijk M, van Buel BL, van Golde LM,Haagsman HP. Surfactant protein composition of lamellar bodies isolatedfrom rat lung. Biochem J 274: 115–119, 1991. doi:10.1042/bj2740115.

30. Pantazi D, Kitsiouli E, Karkabounas A, Trangas T, Nakos G, LekkaME. Dipalmitoyl-phosphatidylcholine biosynthesis is induced by non-injurious mechanical stretch in a model of alveolar type II cells. Lipids 48:827–838, 2013. doi:10.1007/s11745-013-3800-8.

31. Hiansen JQ, Keating E, Aspros A, Yao LJ, Bosma KJ, YamashitaCM, Lewis JF, Veldhuizen RAW. Cholesterol-mediated surfactant dys-function is mitigated by surfactant protein A. Biochim Biophys Acta 1848:813–820, 2015. doi:10.1016/j.bbamem.2014.12.009.

32. Sanders RL, Hassett RJ, Vatter AE. Isolation of lung lamellar bodiesand their conversion to tubular myelin figures in vitro. Anat Rec 198:485–501, 1980. doi:10.1002/ar.1091980310.

33. Schmiedl A, Ochs M, Mühlfeld C, Johnen G, Brasch F. Distribution ofsurfactant proteins in type II pneumocytes of newborn, 14-day old, andadult rats: an immunoelectron microscopic and stereological study. His-tochem Cell Biol 124: 465–476, 2005. doi:10.1007/s00418-005-0066-0.

34. Da Silva K, McCaig LA, Veldhuizen RA, Possmayer F. Protein inhi-bition of surfactant during mechanical ventilation of isolated rat lungs. ExpLung Res 31: 745–758, 2005. doi:10.1080/01902140500267431.

35. Siner JM, Pisani MA. Mechanical ventilation and acute respiratorydistress syndrome in older patients. Clin Chest Med 28: 783–791, 2007.doi:10.1016/j.ccm.2007.08.008.

36. Slutsky AS, Tremblay LN. Multiple system organ failure. Is mechanicalventilation a contributing factor? Am J Respir Crit Care Med 157:1721–1725, 1998. doi:10.1164/ajrccm.157.6.9709092.

37. Tremblay LN, Slutsky AS. Ventilator-induced lung injury: from thebench to the bedside. Intensive Care Med 32: 24–33, 2006. doi:10.1007/s00134-005-2817-8.

38. Valle RP, Wu T, Zuo YY. Biophysical influence of airborne carbonnanomaterials on natural pulmonary surfactant. ACS Nano 9: 5413–5421,2015. doi:10.1021/acsnano.5b01181.

39. Veldhuizen RA, Tremblay LN, Govindarajan A, van Rozendaal BA,Haagsman HP, Slutsky AS. Pulmonary surfactant is altered duringmechanical ventilation of isolated rat lung. Crit Care Med 28: 2545–2551,2000. doi:10.1097/00003246-200007000-00059.

40. Veldhuizen RA, McCaig LA, Akino T, Lewis JF. Pulmonary surfactantsubfractions in patients with the acute respiratory distress syndrome. Am JRespir Crit Care Med 152: 1867–1871, 1995. doi:10.1164/ajrccm.152.6.8520748.

41. Verbrugge SJ, Lachmann B. Mechanisms of ventilation-induced lunginjury: physiological rationale to prevent it. Monaldi Arch Chest Dis 54:22–37, 1999.

42. Vockeroth D, Gunasekara L, Amrein M, Possmayer F, Lewis JF,Veldhuizen RA. Role of cholesterol in the biophysical dysfunction ofsurfactant in ventilator-induced lung injury. Am J Physiol Lung CellMol Physiol 298: L117–L125, 2010. doi:10.1152/ajplung.00218.2009.

44. Welk B, Malloy JL, Joseph M, Yao LJ, Veldhuizen AW. Surfactanttreatment for ventilation-induced lung injury in rats: effects on lungcompliance and cytokines. Exp Lung Res 27: 505–520, 2001. doi:10.1080/019021401750414038.

45. Wurfel MM. Microarray-based analysis of ventilator-induced lunginjury. Proc Am Thorac Soc 4: 77–84, 2007. doi:10.1513/pats.200608-149JG.

46. Zupancich E, Paparella D, Turani F, Munch C, Rossi A, Massaccesi S,Ranieri VM. Mechanical ventilation affects inflammatory mediators inpatients undergoing cardiopulmonary bypass for cardiac surgery: a ran-domized clinical trial. J Thorac Cardiovasc Surg 130: 378–383, 2005.doi:10.1016/j.jtcvs.2004.11.061.

L533LAMELLAR BODIES IN VENTILATION-INDUCED LUNG INJURY

AJP-Lung Cell Mol Physiol • doi:10.1152/ajplung.00055.2017 • www.ajplung.org

by 10.220.32.246 on Septem

ber 7, 2017http://ajplung.physiology.org/

Dow

nloaded from