Role of GADD45a in murine models of radiation- and bleomycin-induced lung injury

Abstract

We previously reported protective effects of GADD45a (growth arrest and DNA damage-inducible gene 45 alpha) in murine ventilator-induced lung injury (VILI) via effects on Akt-mediated endothelial cell signaling. In the present study we investigated the role of GADD45a in separate murine models of radiation- and bleomycin-induced lung injury. Initial studies of wild-type mice subjected to single-dose thoracic radiation (10 Gy) confirmed a significant increase in lung GADD45a expression within 24 h and persistent at 6 wk. Mice deficient in GADD45a (GADD45a(-/-)) demonstrated increased susceptibility to radiation-induced lung injury (RILI, 10 Gy) evidenced by increased bronchoalveolar lavage (BAL) fluid total cell counts, protein and albumin levels, and levels of inflammatory cytokines compared with RILI-challenged wild-type animals at 2 and 4 wk. Furthermore, GADD45a(-/-) mice had decreased total and phosphorylated lung Akt levels both at baseline and 6 wk after RILI challenge relative to wild-type mice while increased RILI susceptibility was observed in both Akt(+/-) mice and mice treated with an Akt inhibitor beginning 1 wk prior to irradiation. Additionally, overexpression of a constitutively active Akt1 transgene reversed RILI-susceptibility in GADD45a(-/-) mice. In separate studies, lung fibrotic changes 2 wk after treatment with bleomycin (0.25 U/kg IT) was significantly increased in GADD45a(-/-) mice compared with wild-type mice assessed by lung collagen content and histology. These data implicate GADD45a as an important modulator of lung inflammatory responses across different injury models and highlight GADD45a-mediated signaling as a novel target in inflammatory lung injury clinically.

Keywords: Akt; GADD45a; bleomycin; lung fibrosis; lung injury; radiation pneumonitis.

Fig. 1.

Radiation-induced increased GADD45a expression in murine lungs. A: GADD45a protein levels detected by Western blotting of lung homogenates from wild-type (WT) mice subjected to single-dose thoracic irradiation (IR, 10 Gy) are shown at various time points (n = 5 animals/condition, *P < 0.05 compared with control). Representative blot shown. Note: image was spliced to show relevant time points. B: immunohistochemical staining for GADD45a (brown) is shown in representative control and irradiated mouse lungs (10 Gy) at the time points shown. Areas of increased staining in bronchiole epithelium and intra-alveolar septae consistent with increased GADD45a expression are indicated by small and large arrows, respectively (n = 3 mice sampled/condition with a minimum of 3 fields sampled from each animal, representative images from individual animals are shown; scale bar = 100 μm).

Fig. 2.

Increased susceptibility of GADD45a^−/− mice to radiation-induced lung injury (RILI). A–C: WT and GADD45a^−/− mice were subjected to low-dose thoracic irradiation (10 Gy) and bronchoalveolar lavage (BAL) fluid was collected at 2 and 4 wk. Compared with WT animals, BAL fluid protein and albumin content as well as total cell counts were significantly increased at 2 wk and further increased at 4 wk (n = 10 animals/group, *P < 0.05). D and E: in parallel studies with the same experimental conditions, BAL fluid cytokines IL-1β and TNF-α were significantly increased in GADD45a^−/− mice after irradiation (10 Gy) at 2 and 4 wk compared with irradiated WT mice at the same time points. BAL cytokine levels were also significantly increased, although to a lesser degree, in irradiated WT animals at both 2 and 4 wk compared with untreated WT controls (n = 10 animals/group, *P < 0.05, †P < 0.05 compared with WT controls).

Fig. 3.

Intermediate RILI susceptibility of GADD45a^+/− mice. WT, GADD45a^+/−, and GADD45a^−/− mice were subjected to high-dose thoracic irradiation (20 Gy) and BAL fluid was collected at 6 wk. BAL protein levels (A) and total cell counts (B) were significantly increased in GADD45a^+/− mice compared with WT animals but were significantly decreased compared with GADD45a^−/− mice at the same time point. BAL cell counts were also significantly increased in GADD45a^+/− mice compared with controls although changes in BAL protein levels were not significant and there was not a significant difference between GADD45a^+/− and GADD45a^−/− animals with respect to protein or cells (n = 5 animals/group, *P < 0.05). C: lung histology demonstrated increased inflammatory cell infiltration in GADD45^+/− mice after irradiation (20 Gy) at 6 wk compared with irradiated WT controls while these changes are even more prominent in irradiated GADD45a^−/− mice at the same time point (n = 3 mice sampled/condition with a minimum of 3 fields sampled from each animal, representative images from individual animals are shown; scale bar = 100 μm).

Fig. 4.

Decreased Akt phosphorylation in RILI-challenged GADD45a^−/− mice. A: increased whole lung total Akt mRNA levels were detected by RT-PCR in RILI-challenged (20 Gy) WT mice at 6 wk (n = 4 animals/group, *P < 0.05). B and C: in separate studies, RILI (20 Gy) was also associated with increased phosphorylated Akt (p-Akt) relative to total Akt by Western blotting of whole lung homogenates at 6 wk compared with untreated WT controls (Ctrl; n = 8 animals/group, *P < 0.05). In contrast, p-Akt/total Akt levels detected by Western blotting of whole lung homogenates were significantly decreased in GADD45a^−/− mice at baseline and at 6 wk after irradiation (20 Gy) compared with their respective WT control animals while there was no difference in p-Akt/total Akt levels in untreated and RILI-challenged GADD45a^−/− mice (n = 5 animals/group, *P < 0.05).

Fig. 5.

Protective effects of Akt in murine RILI. A: lungs from WT mice administered single-dose irradiation (20 Gy) were harvested at various time points (1 day, 4 wk, and 6 wk) and homogenates were used for Western blotting for GADD45a. Representative blot shown. Note: image was spliced to show relevant time points. B and C: Akt^+/− and WT mice were subjected to RILI challenge (20 Gy) and BAL fluid collected for analysis at 6 wk. Compared with WT mice, Akt^+/− mice had significantly increased BAL fluid protein levels and total cell counts (n = 5 animals/group, *P < 0.05, †P < 0.05 compared with respective untreated controls). D and E: in separate studies, WT mice were administered an Akt inhibitor (Akt Inhibitor XIII, Isozyme-Selective, Akti2-1/2, 1 mg/kg) via intraperitoneal (IP) injection 3×/wk beginning 1 wk prior to irradiation (20 Gy) and continuing until collection of BAL fluid 6 wk after irradiation. Compared with WT controls, mice treated with the inhibitor had a significant increase in both BAL protein levels and total cell counts (n = 5 animals/group, *P < 0.05, †P < 0.05 compared with respective untreated controls). F–H: a constitutively active Akt1 mammalian expression vector was used to overexpress Akt in the lungs of WT and GADD45a^−/− mice prior to RILI challenge. Overexpression of Akt was associated with an attenuation of radiation-induced (20 Gy) increases in BAL protein and total cell counts at 6 wk in both WT and GADD45a^−/− animals compared with their respective controls administered vector alone (n = 5 animals/group, *P < 0.05).

Fig. 6.

Increased bleomycin-induced fibrosis in GADD45a^−/− mice. WT and GADD45a^−/− mice were administered bleomycin (0.25 U/kg) via intratracheal (IT) installation. A: assessments of lung injury and fibrosis at 2 wk demonstrate significantly increased lung collagen levels after bleomycin in both groups but with further increase in GADD45a^−/− mice compared with WT controls (n = 5 animals/group, *P < 0.05, †P < 0.05 compared with respective vehicle controls). B: lung histology at 2 wk after bleomycin also confirmed increased inflammatory and fibrotic changes in GADD45^−/− mice compared with WT controls (H&E, hematoxylin and eosin; SR, Sirius red staining for collagen; n = 3 mice sampled/condition with a minimum of 3 fields sampled from each animal, representative images from individual animals are shown; scale bar = 100 μm).

Fig. 7.

Protective effects of Akt in murine bleomycin-induced lung injury. A–B: GADD45a^−/− mice were administered an Akt transgene via IP injection 2×/wk beginning 1 wk prior to bleomycin (1.5 U/kg IT) and continuing until collection of BAL fluid 3 wk after bleomycin. Compared with WT controls, mice treated with the transgene had a significant decrease in both BAL protein levels and total cell counts (n = 5 animals/group, *P < 0.05, †P < 0.05 compared with respective untreated controls). C: overexpression of Akt was also associated with an attenuation of bleomycin-induced (1.5 U/kg IT) increases in BAL collagen levels at 3 wk in GADD45a^−/− animals compared with GADD45a^−/− controls administered vector alone (n = 5 animals/group, *P < 0.05). D: lung histology with the same experimental conditions is shown (n = 3 mice sampled/condition with a minimum of 3 fields sampled from each animal, representative images from individual animals are shown; scale bar = 100 μm).