Therapeutic hypothermia (TH) enhances pulmonary surfactant performance in vivo by molecular mechanisms still unknown. Here, the interfacial structure and the composition of lung surfactant films have been analysed in vitro under TH as well as the molecular basis of its improved performance both under physiological and inhibitory conditions. The biophysical activity of a purified porcine surfactant was tested under slow and breathing-like dynamics by constrained drop surfactometry (CDS) and in the captive bubble surfactometer (CBS) at both 33 and 37 °C. Additionally, the temperature-dependent surfactant activity was also analysed upon inhibition by plasma and subsequent restoration by further surfactant supplementation. Interfacial performance was correlated with lateral structure and lipid composition of films made of native surfactant. Lipid/protein mixtures designed as models to mimic different surfactant contexts were also studied. The capability of surfactant to drastically reduce surface tension was enhanced at 33 °C. Larger DPPC-enriched domains and lower percentages of less active lipids were detected in surfactant films exposed to TH-like conditions. Surfactant resistance to plasma inhibition was boosted and restoration therapies were more effective at 33 °C. This may explain the improved respiratory outcomes observed in cooled patients with acute respiratory distress syndrome and opens new opportunities in the treatment of acute lung injury.
Therapeutic hypothermia (TH) is used as an effective method to limit brain injury after certain types of cardiac arrest and perinatal asphyxia, by reducing patient body temperature from 37 to 33.5 °C1,2. Early studies also suggest some possible benefits in cooled patients suffering of acute respiratory distress syndrome (ARDS)3,4,5,6. ARDS accounts for 10% of all cases in intensive care units worldwide. Mortality remains at 30–40% and no effective treatments are available to date7. Primary ARDS occurs after a direct insult to the lung, whereas secondary ARDS is triggered by an extra-pulmonary pro-inflammatory event7,8,9. Theoretically, TH may be useful to treat primary but might be harmful in secondary ARDS10,11. Conversely, during primary ARDS the injury is at first confined to the lung, resulting in lower compliance and more severe surfactant dysfunction than in the secondary form of the syndrome7,8,9.
During ARDS, surfactant function is impaired in several ways12,13. Protein-rich oedema fluid, high levels of cholesterol and secretory phospholipase A2 (sPLA2) can all inhibit surfactant activity directly and enhance the production of reactive oxygen species (ROS). Lung surfactant is dramatically needed to reduce surface tension under the expiration-mediated shrinking of the alveolar air–liquid interface. The precise proportions of dipalmitoylphosphatidylcholine (DPPC), unsaturated phospholipids and hydrophobic surfactant proteins B (SP-B) and C (SP-C) are essential for its function in a healthy lung12. However, oedema proteins, such as albumin, increase in ARDS patients14 and may compete with surfactant membranes for the air–liquid interface. The resulting interfacial steric barrier affects the in vitro surfactant adsorption and the reduction in surface tension under compression15. Beyond the physiological range, the drastic rise in cholesterol compared to the saturated lipid amount leads to surfactant membranes fluidification12. This impairs the ability of clinical and native surfactants to decrease surface tension upon breathing-like cycles16,17. Besides, methyl-β-cyclodextrin can ex vivo overcome this cholesterol-mediated inhibition in a mice model of ARDS18,19. PLA2 targets and hydrolyses the condensed domains of phosphatidylcholine monolayers with a crumbling effect at the air–liquid interface20. Consistently, high levels of the secretory form of the enzyme was detected in ARDS patients along with changes in surfactant phospholipid composition and surface tension properties21. As a result, surfactant amount and performance are reduced and the work of breathing increases upon ARDS13.
Interestingly, cooled animals with different types of lung failure showed better respiratory mechanics and reduced lung tissue inflammation22,23,24. Moreover, reports on cooled asphyxiated neonates with and without neonatal ARDS due to meconium aspiration describe an improvement in surfactant function25,26,27, pulmonary inflammation27, oxygenation and some clinical outcomes such as reduction of respiratory support and hospital stay25,28. The biological mechanisms underlying this benefit are not totally clear. DPPC turnover did not change in cooled neonates29, but a reduction in sPLA2 activity has been reported30.
However, several biophysical mechanisms may also contribute to surfactant performance under cooling. Drastic changes in body temperature leads to adaptive variations in both surfactant structure and composition to maintain proper interfacial properties in hibernating animals31,32. Besides, the surface activity of rabbit lung extracts tends to improve decreasing temperature from 40 to 15 °C33.
It is well known that the organization and physical properties of lipid membranes, like those in lung surfactant, can be influenced by shifting temperatures towards their melting points34. Cooling to values below 30 °C affects lung surfactant performance in the presence of plasma compounds35. However, the reduction from 37 to 33 °C may have a different effect, as it is still above surfactant melting temperature (≈ 32 °C)34.
To the best of our knowledge, no detailed in vitro studies have investigated surfactant biophysical properties under TH. Here, we analyse for the first time both the interfacial structure and the composition of lung surfactant films subjected to compression at 33 °C compared with 37 °C. We also describe enhanced surfactant function and resistance to inhibition by plasma under cooling, as a well as a more efficient restoration of function as a consequence of further surfactant supplementation. Finally, we investigate the molecular mechanisms governing this enhanced performance and propose a feasible biophysical model.
Moderate hypothermia enhances surfactant activity in a concentration-dependent manner
The activity of purified porcine surfactant (PS) was tested at different concentrations (1.5, 2.5 and 5 mg/mL) under slow (2 cycles/min) and quick (20 cycles/min) dynamic cycles by Constrained Drop Surfactometer (CDS) and Captive Bubble Surfactometer (CBS), respectively (Fig. 1). PS initial adsorption did not change among conditions (Fig. 1a), whereas dynamic activity significantly depended on sample concentration.
At limiting concentration (1.5 mg/mL), PS significantly reduced surface tension after 5 slow cycles at 33 °C, whereas it needed 10 cycles at 37 °C (Fig. 1a,b). This trend was also evident at a PS concentration of 2.5 mg/mL. Using the latter condition, there was a significant difference in γmin comparing cycle 5 and 10 for material tested at 37 °C, but it was not the case for PS assayed at 33 °C (Supplementary Fig. S1a). However, this trend was not present at higher concentration (5 mg/mL) (Supplementary Fig. S1b).
Similar results were obtained in CBS experiments. At limiting concentration, γmin is significantly lower after 10 quick cycles at 33 °C compared with 37 °C. Consistently, the minimum surface tension was significantly higher along cycles at 37 °C. At the same time, the maximum surface tension significantly rose upon the interfacial dynamics irrespective of the experimental temperature (Fig. 1c). This is probably related to the highly diluted surfactant used for the experiments and the resulting easy loss of material from the air–liquid interface under compressions.
Moderate hypothermia increases the size of condensed domains in surfactant films, promoting the exclusion of unsaturated phospholipids upon breathing
The activity and structure of PS were tested in a Langmuir–Blodgett trough (Fig. 2) before and after 10 compression-expansion cycles (0.3 cycles/min). PS activity was slightly better at 33 °C compared with 37 °C. Maximum surface pressures were significantly higher (minimum surface tensions significantly lower) at 33 °C compared with 37 °C. This occurred both before and in the first cycle of the experiments (Fig. 2a,c).
Upon cycling, the plateau during expansions was visible at higher surface pressures at 33 °C than 37 °C (Fig. 2a). This suggested that the compressed phase was more stable at 33 °C while the material tended to relax quicker at physiological temperature. Moreover, under compression of the air–liquid interface, a higher amount of surfactant seems to be excluded at 37 °C than at 33 °C and not replaced during the subsequent expansion. This results in a tendency to increase minimum surface pressures at 33 °C taking the last cycles into account (p = 0.09) (Fig. 2a,c).
Upon compression, and before further cycling, the condensed black domains at the surface films were rounder and less polymorphic than after compression-expansion cycling, as observed under epifluorescence microscopy (Fig. 2b). This result was detected regardless of the temperature, suggesting an intrinsically fluid character of the ordered phase before cycling, with a reduced contact perimeter with the coexisting surrounding disordered phases. On the contrary, a de-mixing of lipids, together with larger gel condensed phases, were observed after cycling, especially at 33 °C. These phases apparently contained smaller domains that were organized in clusters. Moreover, brilliant fluorescent spots were present at both temperatures and remained visible after cycles, especially at 33 °C. These probe-rich spots are likely areas of three-dimensional exclusion that appear bright due to the light scattering. Their apparent size was reduced after cycling associated with a lighter green liquid-expanded phase. Interestingly, due to the lower lipid miscibility at 33 °C than 37 °C, phase-coexistence was more evident under hypothermia temperature. This leads to larger green spots, presumably excluded from the interface, along with larger and darker condensed domains at 33 °C than 37 °C (Fig. 2b, Supplementary Fig. S2).
The composition of PS lipids that were proximally associated to the surface under compression was analysed by LC-HRMS (Supplementary Table S1). The percentage of saturated lipids (especially DPPC) and cholesterol increased at 33 °C compared with 37 °C, whereas the amount of unsaturated lipids, especially those with more than 2 double bonds (polyunsaturated phatidylcholine (puPC)), decreased (Fig. 2d).
DPPC proportion, hydrophobic surfactant proteins and steric hindrance of excluded phospholipids contribute to improve surfactant activity at 33 °C
Several surfactant lipid mixtures (DPPC, palmitoyloleoylphosphatidylglicerol (POPG), palmitoyloleoylphosphatidylcholine (POPC) and dioleoylphosphatidylcholine (DOPC) at different proportions) with decreasing capability to pack under compression were combined with SP-B alone (1% total protein-to-lipid by weight) or with both SP-B and SP-C (2% total protein-to-lipid by weight). The combination of DPPC, POPC and POPG roughly simulates the composition of a typical lung surfactant in terms of saturated/unsaturated and zwitterionic/anionic phospholipid species. The presence of DOPC brings the contribution of phospholipids (PLs) with a large area per molecule, such as it is the case with polyunsaturated species.
The surface properties of each mixture were assayed under quick cycles (20 cycles/min) by CBS (Fig. 3a,b). No differences among 33 and 37 °C were noticed for PS at 8 mg/mL (data not shown). Thus, only results at physiological temperature (37 °C) for PS are shown and considered as control for optimal biophysical activity (γmin < 2mN/m, γmax≈ 30 mN/m).
The presence of both SP-B and SP-C increased the adsorption and dynamic activity of all tested materials regardless of the temperature. Upon dynamic cycling, there was a significant reduction in minimal surface tensions at 33 °C compared with 37 °C when a high percentage of DPPC (65%) with respect to POPG (35%) was present (Fig. 3a, second line). This difference was more evident in the absence of SP-C, suggesting the importance of both hydrophobic proteins to exclude POPG from the interface under compression. Moreover, although both SP-B and SP-C were present (Fig. 3a, second line on the left), a trend to significance was observed at 33 °C for the reduction in area that is required to reach minimal surface tension: [13.6 (2.3)% at 33 °C vs 23.8 (2.2)% at 37 °C, p = 0.05].
Together with the DPPC content, the steric hindrance of excluded PLs under compression also enhanced surfactant activity at 33 °C. The difference among temperatures was significant in the presence of DOPC (Fig. 3a,b, fourth line). Conversely, the lipid-protein mixture containing POPC did not show any temperature-mediated changes in surface-active properties. PLs of this mixture probably did not mix well, leading to a marked variability among replicates in CBS and the co-existence of 2 peaks in Differential Scanning Calorimetry (DSC) (Supplementary Fig. S3a).
Finally, the dynamic properties of each mixture and PS were also compared (Fig. 3b). Regarding minimum surface tensions, PS activity was comparable to the activity of the mixture containing DPPC/POPG (65/35%, w/w) with SP-B alone or with SP-C. This result was more evident for material tested at 33 °C. Interestingly, the melting temperatures of these 2 mixtures (Supplementary Fig. S3a,b) were very close to TH: 32.4 (0.12) and 33.9 (0.06) °C.
Moderate hypothermia decreases surfactant inhibition by plasma and enhances the restoration activity of therapeutic surfactants
The inhibition of PS activity by plasma was tested by CDS upon slow cycling (2 cycles/min) at different surfactant concentrations (1.5, 2.5 and 5 mg/mL), both at 33 and 37 °C (Fig. 4a). Within each condition, no differences between temperatures in initial adsorption and γmax were noticed. Conversely, the dynamic activity (γmin) of PS at 1.5 mg/mL was clearly impaired in the presence of plasma (Fig. 4a, second line). Interestingly, at this concentration, PS capability to produce very low surface tension was reduced with and without plasma at 37 °C, but still functional in the absence of plasma at 33 °C (Fig. 4a, second line). Moreover, when PS was tested at 2.5 mg/mL it worked properly in the presence of plasma at 33 °C, but it was inactivated at 37 °C (Fig. 4a, second line).
Restoration experiments (Supplementary Fig. S4) were then performed, maintaining the same amount of plasma, but using different concentrations of PS depending on the experimental temperature: 1.5 mg/mL at 33 °C and 2.5 mg/mL at 37 °C. The restoration activities of two different materials were studied under these conditions: (1) Poractant α (PORα), a widely used therapeutic surfactant, and (2) DPPC/POPG (65/35, w/w) + SP-B/SP-C (2% total protein-to-lipid by weight), the lipid-protein mixture that had shown the best surface-activity (comparable to PS). Several amounts of these materials were used to mimic a therapy, by adding small volumes at increasing concentrations on top of plasma-inhibited surfactant (Supplementary Table S2, Fig. S4) until reaching restoration. These amounts were calculated as the mass percentage of dispensed PLs to PLs in the inhibited surfactant (e.g. 5 μL of PS at 1.5 μg/μL were used at 33 °C, thus 80% of required therapy means 6 μg of therapy). Since the concentration of surfactant material highly influences its surface-active properties36, we decided to use the same therapy concentration for both temperatures. This was performed by dispensing higher volumes at 37 °C compared with 33 °C, still maintaining values up to 4% of the constrained drop volume. As a result, the same percentage (with respect to total PLs in inhibited PS) and concentration of therapy were used regardless of the experimental temperature. Moreover, due to the different starting PS concentration (1.5 and 2.5 mg/mL for 33 and 37 °C, respectively), we focused on the therapy percentage necessary to rescue PS activity for data interpretation. The effects of adding buffer alone to the inhibited PS were also analysed as a reference (Fig. 4b).
Regardless of the therapy used, the amount necessary to restore PS activity was always lower at 33 °C compared with 37 °C. Moreover, the lipid-protein mixture showed better restoration (80–100%), compared to PORα (200%) with a temperature-mediated improvement in the restoration capability. For this material, the restoration dose required was lower at 33 °C than 37 °C (Fig. 4b). This was probably due to the better exclusion of POPG from the interface under compression at 33 °C, leaving enough DPPC to reduce surface tension. To further investigate this hypothesis, three lipid-protein mixtures with decreasing DPPC/POPG ratios were tested with plasma in CDS at 33 and 37 °C and compared to PORα (Fig. 5a,b). Interestingly, the lipid-protein mixtures with DPPC ≥ 50% and POPG ≤ 50%, seem to be more resistant to plasma inhibition than PORα regardless of the experimental temperature. On the contrary, when increasing the amount of POPG to 65%, the activity was inhibited at 37 °C but still functional at 33 °C (Fig. 5b).
We found that TH induces several effects on surfactant properties, compared with the physiological temperature: (1) the capability of PS to drastically reduce surface tension under breathing-like conditions is enhanced in a concentration-dependent manner; (2) surfactant structure and composition at the air–liquid interface re-organize differently under compression (expiration): larger DPPC-enriched condensed domains and lower percentages of less active lipids are detected at 33 °C; (3) PS resistance to plasma inhibition is boosted under cooling; (4) in vitro restoration therapies are more effective at 33 than at 37 °C; (5) the higher the DPPC proportion, the better the restoration performance under TH.
Up to now, no causal therapies are available for ARDS. Although surfactant inhibition is not necessarily the primary pathogenic factor during ARDS7, the protein-rich oedema fluid leaking into the alveoli and the local inflammation make its inactivation a secondary and critical step of the syndrome13,37,38,39. This occurs through different biological mechanisms. For instance, several plasma proteins inhibit the proper adsorption of surfactant, creating a steric barrier at the interface (albumin, fibrinogen), fluidifying surfactant membranes (C-reactive protein) or degrading surfactant lipids (lipases, sPLA2) and proteins (proteases)12,13,40. Here, we observed that the initial adsorption of diluted PS in the presence of plasma was similar at 33 °C and 37 °C (Fig. 4a). However, plasma compounds were better excluded at 33 °C upon interfacial dynamics. Consistently, the γmin was very low regardless of the plasma presence, testing surfactant at 2.5 mg/mL and 33 °C. On the contrary, there was a significant increase in γmin when PS was assayed with plasma at 37 °C (Fig. 4a, second line).
Interestingly, some plasma proteins, such as fibrinogen, can be found into the liquid-expanded phase during lateral compression41. Due to the lipid de-mixing, we observed larger condensed (DPPC-enriched) domains at 33 °C and a liquid-expanded phase with larger excluded three-dimensional structures (Fig. 2b, Supplementary Fig. S2). A temperature-dependent appearance of condensed domains during interfacial compression was already described for DPPC and bovine surfactant42. Thus, we can speculate that a similar mechanism occurs with plasma compounds. Plasma proteins might be better (at lower pressures) squeezed out from the interface at 33 °C, thus increasing the resistance to plasma inactivation under TH.
Moreover, in both adult and meconial ARDS, high cholesterol and polyunsaturated phosphatidylcholine levels were described as a lung response or due to plasma extravasation and meconium presence25,43,44. This may impede surfactant to properly reduce surface tension during expiration. For instance, increasing the amount of cholesterol with respect to DPPC leads to changes in the proper packing of DPPC domains, affecting the capability to rise surface pressures upon lateral compression45.
Interestingly, the surface elasticity of spread DPPC monolayers at high surface pressures seems to lower under increasing temperatures46. This reduces the interfacial activity of material, probably due to the quicker relaxation of DPPC monolayers that lose their capability to sustain high surface pressure. In our experiments, an increased percentage of cholesterol and saturated lipids was detected at the air–liquid interface at 33 °C (Fig. 6).
The sterol presence seems to influence the dynamic surface elasticity of DPPC monolayers under compression47. However, we still observed a temperature-mediated effect on the lateral organization of DPPC/unsaturated PC domains of PS at the air–liquid interface. This is probably due to the simultaneous increase in both cholesterol and saturated lipids detected at hypothermia temperature. In this regard, we demonstrated formation of larger condensed interfacial domains at 33 °C than at 37 °C during lateral compression (Figs. 2b, 6, Supplementary Fig. S2). Cholesterol may insert preferentially into ordered structures. Thus, a larger DPPC proportion at the interface at 33 °C (Fig. 2d) might cause a higher amount of cholesterol to partition from the surrounding disordered phases. These results may suggest a protective role of TH, reducing the cholesterol-mediated inhibition. This might occur by promoting DPPC de-mixing during compression and indirectly the sterol captures into ordered domains at 33 °C.
During ARDS, raised levels of sPLA2 leads to a significant reduction in both surfactant DPPC13,37,38,39 and PG amounts21. At the same time, surfactant proteins decrease13 and ROS increase, oxidizing the lipid-protein membranes12. As a result, the availability of surface-active material is drastically reduced and surfactant complexes become inactivated. Interestingly, considering our data, our findings suggest that TH could be also beneficial under these impaired surfactant conditions: (1) at lower concentration, as in the ARDS, surfactant performance is better at 33 °C than 37 °C (Fig. 1), (2) surfactant exhibits higher resistance to plasma inhibition under TH (Fig. 4); (3) due to the lower miscibility of lipids at 33 °C, unsaturated lipids (e.g. puPC and DOPC) and plasma components are removed more efficiently from the interface. This occurs with a lower loss of DPPC during the interfacial compression (at expiration) (Figs. 2, 5). This last mechanism (outlined in Fig. 7) fits with the first two effects perfectly. A better performance of surfactant at 33 °C with and without plasma is closely linked to the concentration of tested material. Lower concentrations mean lower DPPC amounts at the air–liquid interface, that are even more reduced under compression at 37 °C.
In asphyxiated neonates, the improvement in surfactant activity under TH seems to be significant after 48–72 h25. At this time point, two sPLA2 IIA modulators, SP-A and DOPG, might increase and the activity of the enzyme decreases30. This may explain the time-dependent effect on surfactant activity under TH, probably related to the higher DPPC amount at the interface.
Finally, we demonstrated an increase in both the endogenous and therapeutic surfactant performance under cooling (Figs. 4, 5). Regardless of the material applied (PORα or lipid-protein mixtures containing different proportions of DPPC and POPG), lower doses were more resistant to plasma inhibition and able to restore PS activity at 33 °C compared with 37 °C. Moreover, the resistance to plasma inactivation was higher at high DPPC ratios regardless of the temperature (Fig. 5). Altogether, these results are promising for primary ARDS patients. In those patients, the maximum beneficial effect of exogenous surfactant should be warranted with the minimum dose.
Our data show a better plasma resistance for the lipid-protein mixture containing ≥ 50% DPPC when compared to PORα at both 33 °C and 37 °C (Fig. 5). Similar observations were recently described for CHF 5633 at physiological temperature, a novel synthetic surfactant that shows promising results in clinical phase II48,49. CHF 5633 composition is similar to the lipid-protein mixtures employed in our study: 50/50 DPPC/POPG (w/w%) and 1.8% by mass of SP-B/SP-C analogs48. Thus, the mechanism involved in plasma resistance upon dynamic cycles may probably be analogous and particularly enhanced at 33 °C. In this line, the lipid-protein mixture with increased POPG/DPPC ratio showed a reduced activity at 37 °C, but was still functional at 33 °C. We speculate that plasma proteins are excluded from the interface together with POPG during compression, presumably in a process lead by surfactant proteins/peptides. The same mechanism might occur under TH, with less DPPC lost from the interface during compression (Fig. 7). The lower lipid miscibility at 33 °C may contribute to reduce DPPC mixing into the liquid-expanded phase and its interfacial exclusion under compression. This would explain the good performance of the material at low DPPC ratio (35%) (Fig. 5).
We acknowledge some study limitations. PS undergoes a temperature-mediated collapse under compression in the Langmuir–Blodgett balance50. This prevents analysing surfactant interfacial structure and composition at surface pressures higher than 40 mN/m. In addition, we could not perform a lipidomic analysis after compression-expansion cycles since these experiments last around 30 min and the lipids at the interface may oxidize during this period. Also, experiments using lipid-protein mixtures containing only SP-C could not be carried out in the CBS, because the presence of SP-B is indeed essential for the adsorption and re-adsorption of a significant proportion of lipids at the bubble interface51. As previously described, surfactant is resistant to plasma inactivation under quick cycles in the CDS52. For that reason, the use of slow cycles was necessary to facilitate the competition of plasma proteins for the air–liquid interface. Finally, we cannot exclude that the lower interfacial activity at 37 °C, especially for the mixture containing high percentages of POPG or DOPC, was influenced by a higher lipid oxidation at increased temperatures53.
Altogether, we demonstrated that TH influences surfactant structure and composition at the air–liquid interface, increasing the interfacial proportion of DPPC upon compression. This leads to improved lung and therapeutic surfactant activity both under physiological and inhibitory conditions. Our evidences should be added to the multiple effects of TH on human lungs.
PS and purified SP-B and C were obtained from porcine bronchoalveolar lavage (BAL) as previously reported15 (Supplementary Fig. S5). The experimental procedures are described in the supplementary information. Porcine plasma was obtained by centrifuging (500×g for 10 min at 4 °C) and pooling blood from 6 pigs with different sexes coming from the same slaughterhouse. Similarly, PS was purified from the BAL pool obtained from the 6 pig lungs. Animals used to recover these materials were healthy and subjected to Vet control according to local regulations (Spanish hygiene rules legislation, law articles 83, 85 and 178). Pigs were sacrificed for food and not for the sole purpose of the study.
PORα is a widely available surfactant obtained by minced porcine lung and considered to be the most efficacious in neonatal critical care, according to evidence based data and European guidelines56,57. It was purchased from Chiesi Farmaceutici S.p.A. (Parma, Italy), lyophilized and stored at − 80 °C until use. DPPC, POPG, POPC and DOPC were purchased from Avanti Polar Lipids, Inc. (Alabaster, Alabama, USA). BODIPY-PC was purchased from Molecular Probes (Life Technologies, Carlsbad, California, USA). The same buffer solution containing 5 mM Tris and 150 mM NaCl at pH 7.4 was used for every experiment.
Biophysical activity and lateral structure of surfactant films
The biophysical activity of both PS and lipid-protein surfactant suspensions was tested at limiting concentrations (1.5, 2, 5 and 8 mg/mL) under slow (2 cycles/min) and quick breathing-like (20 cycles/min) compression-expansion cycling by CDS58 and CBS51 (Supplementary Fig. S6). Surfactant capability to reduce surface tension was not inhibited during quick cycles in the CDS after bulk replacement with serum52. Thus, both plasma inhibition and therapy restoration experiments were performed under slow cycles (Supplementary Fig. S6, Table S2). A Langmuir–Blodgett surface balance was used to study PS lateral structure (Supplementary Fig. S8) before and after subjecting material (around 30 μg at 5 mg/mL) at the interface to 10 compression-expansion cycles (65 cm2/min).
All methods were applied keeping the target temperature (33 °C or 37 °C) constant along the experiments. Details about protocols and techniques are described in supplementary information.
Composition of interfacial films and thermotropic properties
The temperature-dependent changes in lipid composition of PS films under compression at the air–liquid interface were analysed in a Langmuir–Blodgett balance (Supplementary Fig. S8). Briefly, PS (around 50 μg, 5 mg/mL), pre-heated to 33 °C or 37 °C, was deposited drop by drop at the air–liquid interface of the trough filled with buffer at the same temperature. After 10 min of equilibration (≈ 25 mN/m surface pressure), surfactant films were compressed, reducing interfacial area at a velocity of 25 cm2/min and simultaneously transferred onto a glass slide. Immediately after drying, the glass slide was rinsed with chloroform/methanol (2:1 v/v) to collect lipids from the film. Five experiments per condition were pooled in a single sample for lipidomic analysis by liquid chromatography-high resolution mass spectrometry (LC-HRMS)43. Each lipid class was expressed as molar % of the total PLs.
DSC was performed on both PS and lipid/protein suspensions, testing materials at 3 mg/mL as described earlier48 and in supplementary information.
Data were expressed as mean (standard deviation). Results comparisons were conducted using One-Way ANOVA followed by post-hoc Tukey’s test or Two-Way ANOVA followed by post-hoc paired and unpaired T-test when appropriate. Details are described in figure captions. Correlation analysis was performed using Spearman (ρ) coefficient. Analyses were carried out with Sigma Plot (v. 11, Systat Software, San Jose, USA) and IBM SPSS Statistics 25 (v.25, IBM Corp., Armonk, NY, USA). p < 0.05 was considered to be significant.
Arrich, J., Holzer, M., Havel, C., Mullner, M. & Herkner, H. Hypothermia for neuroprotection in adults after cardiopulmonary resuscitation. Cochrane Database Syst. Rev. 2, CD004128. https://doi.org/10.1002/14651858.CD004128.pub4 (2016).
Pietrini, D. et al. Neuroprotection and hypothermia in infants and children. Curr. Drug Targets 13, 925–935. https://doi.org/10.2174/138945012800675641 (2012).
Hayek, A. J., White, H. D., Ghamande, S., Spradley, C. & Arroliga, A. C. Is therapeutic hypothermia for acute respiratory distress syndrome the future?. J. Intens. Care Med. 32, 460–464 (2017).
Slack, D. F. et al. Pilot feasibility study of therapeutic hypothermia for moderate to severe acute respiratory distress syndrome. Crit. Care Med. 45, 1152 (2017).
White, H. D., Ghamande, S. & Arroliga, A. C. Therapeutic hypothermia for acute respiratory distress syndrome. Crit. Care Med. 45, e1202–e1203 (2017).
Karnatovskaia, L. V., Festic, E., Freeman, W. D. & Lee, A. S. Effect of therapeutic hypothermia on gas exchange and respiratory mechanics: a retrospective cohort study. Therap. Hypothermia Temp. Manage. 4, 88–95 (2014).
Matthay, M. A. et al. Acute respiratory distress syndrome. Nat. Rev. Dis. Primers 5, 1–22 (2019).
Calfee, C. S. et al. Distinct molecular phenotypes of direct vs indirect ARDS in single-center and multicenter studies. Chest 147, 1539–1548. https://doi.org/10.1378/chest.14-2454 (2015).
Sinha, P. & Calfee, C. S. Phenotypes in acute respiratory distress syndrome: moving towards precision medicine. Curr. Opin. Crit. Care 25, 12–20. https://doi.org/10.1097/mcc.0000000000000571 (2019).
Itenov, T. S. et al. Induced hypothermia in patients with septic shock and respiratory failure (CASS): a randomised, controlled, open-label trial. Lancet Respir. Med. 6, 183–192. https://doi.org/10.1016/s2213-2600(18)30004-3 (2018).
Rumbus, Z. et al. Fever is associated with reduced, hypothermia with increased mortality in septic patients: a meta-analysis of clinical trials. PLoS ONE 12, e0170152. https://doi.org/10.1371/journal.pone.0170152 (2017).
Autilio, C. & Pérez-Gil, J. Understanding the principle biophysics concepts of pulmonary surfactant in health and disease. Arch. Dis. Childhood-Fetal Neonatal Ed. 104, F443–F451 (2019).
Günther, A. et al. Surfactant alteration and replacement in acute respiratory distress syndrome. Respir. Res. 2, 353 (2001).
Sprung, C. L., Rackow, E. C., Fein, I. A., Jacob, A. I. & Isikoff, S. K. The spectrum of pulmonary edema: differentiation of cardiogenic, intermediate, and noncardiogenic forms of pulmonary edema. Am. Rev. Respir. Dis. 124, 718–722 (1981).
Taeusch, H. W., de le Serna, J. B., Perez-Gil, J., Alonso, C. & Zasadzinski, J. A. Inactivation of pulmonary surfactant due to serum-inhibited adsorption and reversal by hydrophilic polymers: experimental. Biophys. J. 89, 1769–1779. https://doi.org/10.1529/biophysj.105.062620 (2005).
Keating, E. et al. Effect of cholesterol on the biophysical and physiological properties of a clinical pulmonary surfactant. Biophys. J. 93, 1391–1401. https://doi.org/10.1529/biophysj.106.099762 (2007).
López-Rodríguez, E., Ospina, O. L., Echaide, M., Taeusch, H. W. & Pérez-Gil, J. Exposure to polymers reverses inhibition of pulmonary surfactant by serum, meconium, or cholesterol in the captive bubble surfactometer. Biophys. J. 103, 1451–1459 (2012).
Gunasekara, L. C. et al. Methyl-beta-cyclodextrin restores the structure and function of pulmonary surfactant films impaired by cholesterol. Biochim. Biophys. Acta 1798, 986–994. https://doi.org/10.1016/j.bbamem.2009.12.003 (2010).
Al-Saiedy, M. et al. Surfactant dysfunction in ARDS and bronchiolitis is repaired with cyclodextrins. Mil. Med. 183, 207–215. https://doi.org/10.1093/milmed/usx204 (2018).
Grainger, D., Reichert, A., Ringsdorf, H. & Salesse, C. An enzyme caught in action: direct imaging of hydrolytic function and domain formation of phospholipase A2 in phosphatidylcholine monolayers. FEBS Lett. 252, 73–82 (1989).
Seeds, M. C. et al. Secretory phospholipase A2-mediated depletion of phosphatidylglycerol in early acute respiratory distress syndrome. Am. J. Med. Sci. 343, 446–451. https://doi.org/10.1097/MAJ.0b013e318239c96c (2012).
Cruces, P. et al. Mild hypothermia increases pulmonary anti-inflammatory response during protective mechanical ventilation in a piglet model of acute lung injury. Pediatr. Anesth. 23, 1069–1077 (2013).
Hong, S.-B. et al. Induced hypothermia as a new approach to lung rest for the acutely injured lung. Crit. Care Med. 33, 2049–2055 (2005).
Ball, M. K. et al. Body temperature effects on lung injury in ventilated preterm lambs. Resuscitation 81, 749–754 (2010).
Autilio, C., Echaide, M., De Luca, D. & Pérez-Gil, J. Controlled hypothermia may improve surfactant function in asphyxiated neonates with or without meconium aspiration syndrome. PLoS ONE 13, e0192295. https://doi.org/10.1371/journal.pone.0192295 (2018).
Autilio, C., Echaide, M., Dell’Orto, V., Perez-Gil, J. & De Luca, D. Effect of whole body hypothermia on surfactant function when amniotic fluid is meconium stained. Ther. Hypothermia Temp. Manage. 10, 186–189. https://doi.org/10.1089/ther.2017.0012 (2020).
De Luca, D. et al. Effect of whole body hypothermia on inflammation and surfactant function in asphyxiated neonates. Eur. Respir. J. 44, 1708–1710 (2014).
De Luca, D. et al. Hypothermia and meconium aspiration syndrome: international multicenter retrospective cohort study. Am. J. Respir. Crit. Care Med. 194, 381–384 (2016).
Nespeca, M. et al. Does whole-body hypothermia in neonates with hypoxic-ischemic encephalopathy affect surfactant disaturated-phosphatidylcholine kinetics?. PLoS ONE 11, e0153328 (2016).
Autilio, C., Shankar-Aguilera, S., Minucci, A., Touqui, L. & De Luca, D. Effect of cooling on lung secretory phospholipase A2 activity in vitro, ex vivo, and in vivo. Am. J. Physiol.-Lung Cell. Mol. Physiol. 316, L498–L505 (2019).
Suri, L. N. et al. Adaptation to low body temperature influences pulmonary surfactant composition thereby increasing fluidity while maintaining appropriately ordered membrane structure and surface activity. Biochim. et Biophys. Acta Biomembr. 1818, 1581–1589 (2012).
Suri, L. N. et al. Adaptations to hibernation in lung surfactant composition of 13-lined ground squirrels influence surfactant lipid phase segregation properties. Biochim. et Biophys. Acta Biomembr. 1828, 1707–1714 (2013).
Lempert, J. & Macklem, P. T. Effect of temperature on rabbit lung surfactant and pressure-volume hysteresis. J. Appl. Physiol. 31, 380–385. https://doi.org/10.1152/jappl.1918.104.22.1680 (1971).
de la Serna, J. B., Perez-Gil, J., Simonsen, A. C. & Bagatolli, L. A. Cholesterol rules direct observation of the coexistence of two fluid phases in native pulmonary surfactant membranes at physiological temperatures. J. Biol. Chem. 279, 40715–40722 (2004).
Enhorning, G., Hohlfeld, J., Krug, N., Lema, G. & Welliver, R. Surfactant function affected by airway inflammation and cooling: possible impact on exercise-induced asthma. Eur. Respir. J. 15, 532–538 (2000).
Gunasekara, L., Schoel, W. M., Schürch, S. & Amrein, M. W. A comparative study of mechanisms of surfactant inhibition. Biochim. et Biophys. Acta Biomembr. 1778, 433–444 (2008).
Gunther, A. et al. Surfactant alterations in severe pneumonia, acute respiratory distress syndrome, and cardiogenic lung edema. Am. J. Respir. Crit. Care Med. 153, 176–184. https://doi.org/10.1164/ajrccm.153.1.8542113 (1996).
Schmidt, R. et al. Time-dependent changes in pulmonary surfactant function and composition in acute respiratory distress syndrome due to pneumonia or aspiration. Respir. Res. 8, 55. https://doi.org/10.1186/1465-9921-8-55 (2007).
Schmidt, R. et al. Alteration of fatty acid profiles in different pulmonary surfactant phospholipids in acute respiratory distress syndrome and severe pneumonia. Am. J. Respir. Crit. Care Med. 163, 95–100. https://doi.org/10.1164/ajrccm.163.1.9903029 (2001).
Echaide, M., Autilio, C., Arroyo, R. & Perez-Gil, J. Restoring pulmonary surfactant membranes and films at the respiratory surface. Biochim. Biophys. Acta Biomembr. 1725–1739, 2017. https://doi.org/10.1016/j.bbamem.2017.03.015 (1859).
Williams, I., Zasadzinski, J. A. & Squires, T. M. Interfacial rheology and direct imaging reveal domain-templated network formation in phospholipid monolayers penetrated by fibrinogen. Soft Matter 15, 9076–9084. https://doi.org/10.1039/c9sm01519a (2019).
Discher, B. M. et al. Lateral phase separation in interfacial films of pulmonary surfactant. Biophys. J. 71, 2583–2590. https://doi.org/10.1016/s0006-3495(96)79450-x (1996).
Autilio, C. et al. Surfactant injury in the early phase of severe meconium aspiration syndrome. Am. J. Respir. Cell Mol. Biol. 63, 327–337. https://doi.org/10.1165/rcmb.2019-0413OC (2020).
Markart, P. et al. Patients with ARDS show improvement but not normalisation of alveolar surface activity with surfactant treatment: putative role of neutral lipids. Thorax 62, 588–594. https://doi.org/10.1136/thx.2006.062398 (2007).
Miyoshi, T. & Kato, S. Detailed analysis of the surface area and elasticity in the saturated 1,2-diacylphosphatidylcholine/cholesterol binary monolayer system. Langmuir 31, 9086–9096. https://doi.org/10.1021/acs.langmuir.5b01775 (2015).
Bykov, A. G. et al. Influence of temperature on dynamic surface properties of spread DPPC monolayers in a broad range of surface pressures. Chem. Phys. Lipids 225, 104812. https://doi.org/10.1016/j.chemphyslip.2019.104812 (2019).
Vrânceanu, M., Winkler, K., Nirschl, H. & Leneweit, G. Surface rheology and phase transitions of monolayers of phospholipid/cholesterol mixtures. Biophys. J. 94, 3924–3934. https://doi.org/10.1529/biophysj.107.104851 (2008).
Echaide, M., Autilio, C., López-Rodríguez, E., Cruz, A. & Pérez-Gil, J. In vitro functional and structural characterization of a synthetic clinical pulmonary surfactant with enhanced resistance to inhibition. Sci. Rep. 10, 1–10 (2020).
Ramanathan, R. et al. Synthetic surfactant CHF5633 compared with poractant alfa in the treatment of neonatal respiratory distress syndrome: a multicenter, double-blind, randomized, controlled clinical trial. J. Pediatrics 225, 90 (2020).
Rugonyi, S., Biswas, S. C. & Hall, S. B. The biophysical function of pulmonary surfactant. Respir. Physiol. Neurobiol. 163, 244–255. https://doi.org/10.1016/j.resp.2008.05.018 (2008).
Schürch, D., Ospina, O. L., Cruz, A. & Pérez-Gil, J. Combined and independent action of proteins SP-B and SP-C in the surface behavior and mechanical stability of pulmonary surfactant films. Biophys. J. 99, 3290–3299 (2010).
Saad, S. M. et al. A double injection ADSA-CSD methodology for lung surfactant inhibition and reversal studies. Colloids Surf. B 73, 365–375 (2009).
Ghnimi, S., Budilarto, E. & Kamal-Eldin, A. The new paradigm for lipid oxidation and insights to microencapsulation of Omega-3 fatty acids. Comprehen. Rev. Food Sci. Food Saf. 16, 1206–1218 (2017).
Rouser, G., Siakotos, A. & Fleischer, S. Quantitative analysis of phospholipids by thin-layer chromatography and phosphorus analysis of spots. Lipids 1, 85–86 (1966).
Lowry, O. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265–275 (1951).
Sweet, D. G. et al. European Consensus Guidelines on the management of respiratory distress syndrome—2019 update. Neonatology 115, 432–450. https://doi.org/10.1159/000499361 (2019).
Foligno, S. & De Luca, D. Porcine versus bovine surfactant therapy for RDS in preterm neonates: pragmatic meta-analysis and review of physiopathological plausibility of the effects on extra-pulmonary outcomes. Respir. Res. 21, 8. https://doi.org/10.1186/s12931-019-1267-8 (2020).
Yu, L. M. et al. Constrained sessile drop as a new configuration to measure low surface tension in lung surfactant systems. J. Appl. Physiol. 97, 704–715 (2004).
Authors are grateful to José Carlos Castillo-Sánchez, Olga Cañadas and Alejandro Alonso Eugenio (Complutense University, Madrid) for their technical assistance during the analysis of DSC and epifluorescence experiments.
The study was funded by the Spanish Ministry of Science and Innovation (RTI2018-094564-B-I00) and from the regional Government of Madrid (P2018/NMT-4389) (to CA, CGM, AH, AC, ME and JPG).
AA/Prof.De Luca has received research and educational grants from Chiesi Farmaceutici spa and ABBVIE inc and THERABRON inc.. He served as consultant and lecturer for Airway Therapeutics, Chiesi Pharmaceuticals spa and ABBVIE inc. Finally, he has been member of advisory boards for Chiesi Pharmaceuticals spa and ABBVIE inc. Prof. Perez-Gil has received research grants from Chiesi Farmaceutici spa and Airway Therapeutics. C.A., M.E., A.C., C. G.-M., A.H., E. D.S. and J.B.S. declare no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Autilio, C., Echaide, M., Cruz, A. et al. Molecular and biophysical mechanisms behind the enhancement of lung surfactant function during controlled therapeutic hypothermia. Sci Rep 11, 728 (2021). https://doi.org/10.1038/s41598-020-79025-3