Effect of hypoxia on lung gene expression and proteomic profile: insights into the pulmonary surfactant response

Abstract

Exposure of lung to hypoxia has been previously reported to be associated with significant alterations in the protein content of bronchoalveolar lavage (BAL) and lung tissue. In the present work we have used a proteomic approach to describe the changes in protein complement induced by moderate long-term hypoxia (rats exposed to 10% O2 for 72h) in BAL and lung tissue, with a special focus on the proteins associated with pulmonary surfactant, which could indicate adaptation of this system to limited oxygen availability. The analysis of the general proteomic profile indicates a hypoxia-induced increase in proteins associated with inflammation both in lavage and lung tissue. Analysis at mRNA and protein levels revealed no significant changes induced by hypoxia on the content in surfactant proteins or their apparent oligomeric state. In contrast, we detected a hypoxia-induced significant increase in the expression and accumulation of hemoglobin in lung tissue, at both mRNA and protein levels, as well as an accumulation of hemoglobin both in BAL and associated with surface-active membranes of the pulmonary surfactant complex. Evaluation of pulmonary surfactant surface activity from hypoxic rats showed no alterations in its spreading ability, ruling out inhibition by increased levels of serum or inflammatory proteins.

Biological significance: This work reveals that hypoxia induces extensive changes in the proteomic profile of lung bronchoalveolar lavage, including the presence of proteins related with inflammation both in lung tissue and lavage, and a significant increase in the synthesis and secretion by the lung tissue of different forms of hemoglobin. The level of specific pulmonary surfactant-associated proteins is not substantially altered due to hypoxia, but hypoxia-adapted surfactant exhibits an enhanced ability to form surface-active films at the air-liquid interface. The increased amount of β-globin integrated into the operative surfactant complexes obtained from hypoxic rats is a relevant feature that points to the existence of adaptive responses coupling surfactant function and oxygen availability.

Keywords: ABCA3; Air–liquid interface; Hemoglobin; Lung surfactant; Surfactant proteins.

Figure 1. Validation of a moderate long-term hypoxia model (10% oxygen exposure for 72 hours)

A. Induced mRNA expression of HIF-1α and HIMF hypoxia factors in lung tissue upon hypoxia exposure, as determined by real-time PCR. B. Principal component analysis of proteomic data revealed a clear separation between control and hypoxia groups. Dots represent the 6 animals studied in each group.

Figure 2. Effect of hypoxia in phospholipid and protein content of surfactant and BAL

A. Balance of phospholipids in the small (SA) with respect to the large (LA) surfactant aggregates and total lipid content in pulmonary surfactant from control and hypoxic rats. B. Total protein content in bronchoalveolar lavage (BAL) and purified surfactant, as well as the relative protein to lipid ratio in surfactant, as determined in BAL and surfactant from normoxic and hypoxic rats. Values are presented as means with standard error of the mean (SEM) (n=6 for each group), and significant differences are shown as * (p<0.05).

Figure 3. Effect of hypoxia on the mRNA expression and protein levels of surfactant-related proteins

A. Comparison of the relative expression of the genes coding for pulmonary surfactant proteins A, B, C and D and the lipid transporter ABCA3 involved in surfactant biogenesis, in control and hypoxic lung tissue, as determined by real-time PCR. Values are presented as means with standard error of the mean (SEM) (n=6 for each group). B. Comparison of the relative content of the main surfactant apolipoproteins as determined by Western blot of samples from bronchoalveolar lavage (BAL). Values are presented as means with SEM (n=4 gels quantified by densitometry, with 6 rats from each group per gel).

Figure 4. Changes in mRNA and protein levels of hemoglobin in response to hypoxia

A. Comparison of the relative expression of globin genes in control and hypoxic lung tissue, as determined by real-time PCR. Values are presented as means with standard error of the mean (SEM) (n=6 for each group) and significant differences are shown as * (p<0.05). B. Comparison of the relative content of hemoglobin as determined by Western blot of samples from bronchoalveolar lavage (BAL) and pulmonary surfactant (PS). Values are presented as means with SEM (n=4 gels quantified by densitometry of 14 kDa bands, with 6 rats from each group per gel). Significant differences are shown as * (p<0.05).

Figure 5. Effect of hypoxia on the profile of α-globin isoforms in lung tissue

A. Detail of a reference 2D-DIGE gel from lung tissue samples, showing the spots identified as different α-globin forms (4a, b, c). B. Differences between tissue samples from normoxic and hypoxic animals with respect to the levels of the different α-globin forms, determined by the fluorescent intensity of the single spots. Globin forms that significantly increased under hypoxia are labeled in blue, and those with reduced amount, in red.

Figure 6. Effect of hypoxia on the interfacial spreading and adsorption ability of pulmonary surfactant

Plot of the maximum surface pressure achieved upon spreading of increasing amounts of pulmonary surfactant obtained from control normoxic (open circles) and hypoxic (closed circles). Data are presented as means with SEM (n=3). Significant differences are shown as * (p<0.05).

Figure 7. Profile of hypoxia-affected lung proteome

The diagrams represent the proteins exhibiting a significant content change as a consequence of hypoxia in their respective functional groups in BAL (A) and lung tissue (B). Group overlaps include proteins that are simultaneously involved in several of the functional groups. The numbers correspond to the proteins listed in supplementary Table 3 (reduced level in red, increased in blue). The size of the number font intended to reflect the extent of the change in protein level, according to the data in Tables 1 and 2.