Priming With Endotoxin Increases Acute Lung Injury in Mice by Enhancing the Severity of Lung Endothelial Injury
Version of Record online: 16 NOV 2010
Copyright © 2010 Wiley-Liss, Inc.
The Anatomical Record
Volume 294, Issue 1, pages 165–172, January 2011
How to Cite
Lee, J. S., Su, X., Rackley, C., Matthay, M. A. and Gupta, N. (2011), Priming With Endotoxin Increases Acute Lung Injury in Mice by Enhancing the Severity of Lung Endothelial Injury. Anat Rec, 294: 165–172. doi: 10.1002/ar.21244
- Issue online: 13 DEC 2010
- Version of Record online: 16 NOV 2010
- Manuscript Accepted: 23 MAY 2010
- Manuscript Received: 27 JAN 2010
- NHLBI. Grant Number: HL51854
- NIAID. Grant Number: P011053194
- NHLBI. Grant Number: F32 HL097383
- NHLBI. Grant Number: K08 HL092059
- Parker B. Francis Awards
- acute lung injury;
Endotoxin-induced acute lung injury (ALI) is a commonly used model. However, the effect of a priming dose of endotoxin on lung fluid balance has not been well studied. We hypothesized that endotoxin-induced ALI in mice would be enhanced under a priming condition. Mice were intratracheally (IT) instilled with either a priming dose of endotoxin from E. coli (0.5 mg/kg) or equal volume of PBS. Eighteen hours later, a larger challenge dose of endotoxin (5 mg/kg) was given IT. Control mice received PBS only. After 24 hr, the mice were sacrificed and the degree of lung injury and inflammation were measured. Endotoxin priming increased body weight loss and worsened hypothermia. Extravascular lung water and lung endothelial permeability were higher in the primed group. Priming with endotoxin reduced alveolar fluid clearance; however, there was no effect on bronchoalveolar lavage (BAL) levels of receptor for advanced glycation end products (RAGE). The primed group had increased alveolar inflammation as demonstrated by increased numbers of neutrophils in the BAL. There was no significant difference in NF-κB p65 in the lung nuclear extract among the experimental groups. Taken together, priming with a small dose of endotoxin followed by a larger challenge dose of endotoxin induces more systemic illness and increased pulmonary edema in mice, largely due to increased lung endothelial permeability and lung inflammation. This model should be useful to investigators studying ALI who want to simulate the clinical setting in which more than one insult often leads to greater clinical lung injury. Anat Rec, 2010. © 2010 Wiley-Liss, Inc.
Acute lung injury (ALI) and the acute respiratory distress syndrome (ARDS) affect more than 190,000 persons every year with a mortality rate of 30%–40%, accounting for ∼75,000 deaths per year (Rubenfeld et al.,2005). ALI and ARDS are clinically characterized by acute hypoxemic respiratory failure, bilateral patchy infiltrates, and normal cardiac filling pressures (Ware and Matthay,2000). The causes of ALI/ARDS are heterogeneous and predominantly include sepsis, pneumonia, blood transfusions, aspiration, and trauma (Ware and Matthay,2000). Animal models of ALI have been studied to better understand the mechanisms and pathogenesis of ALI to develop novel therapies for this syndrome.
Endotoxin has been used in several different animal models of ALI (Matute-Bello et al.,2008). The traditional endotoxin model of ALI is commonly performed as a “single-hit” model and leads to activation of NF-κB-dependent pathways (Zhang and Ghosh,2000). However, this single-hit experimental model has been limited, because it does not mimic the combination of insults seen in the human syndrome of ALI/ARDS. The theory of multiple insults leading to the clinical syndrome of ALI/ARDS is logical and stems from experiences at the bedside (Lang and Hickman-Davis,2005). The most notable example of a “two-hit” phenomenon is transfusion-related ALI (Looney et al.,2006,2009; Triulzi,2009).
The model we describe in this study is novel and incorporates a two-hit model of endotoxin-induced ALI. The model involves delivering a smaller “priming” dose of endotoxin into the airspaces of the mouse lung followed by a larger challenge dose of endotoxin. The objective of this study was to describe the phenotype of this priming model to provide investigators a useful, more clinically relevant mouse model to study ALI.
MATERIALS AND METHODS
C57BL/6 male mice (range, 8–10 weeks old; The Jackson Laboratory) were used in all experiments. Animals were maintained in the animal facility at the University of California (San Francisco, CA; UCSF). All experimental protocols were approved by the Institutional Animal Care and Use Committee at UCSF.
Mice were first anesthetized by receiving an intraperitoneal (IP) injection of mixture of ketamine (90 mg/kg) and xylazine (10 mg/kg). Then, endotoxin priming was induced by an intratracheal (IT) instillation of lipopolysaccharide (LPS) from E. coli O55:B5 (Sigma) at 0.5 mg/kg or the equivalent volume of PBS. Mice were then allowed to recover in a 100% oxygen chamber for 2 hr while awakening from anesthesia. Finally, 18 hr after the priming, the mice were reanesthetized and given either an IT challenge dose of endotoxin (5 mg/kg) or an equivalent volume of PBS. The mice recovered in the oxygen chamber and followed for 24 hr to measure body weight and temperature in three groups: (1) the control group: PBS IT priming + PBS IT challenge; (2) the unprimed group: PBS IT priming + LPS IT challenge, and (3) the primed group: LPS IT priming + LPS IT challenge. Mice in these three groups, except those analyzed for measurement of NF-κB p65 in the lung nuclear extract, were euthanized at 24 hr after IT LPS or PBS challenge.
Intratracheal Priming and Challenge of LPS
As described previously (Su et al.,2004; Gupta et al.,2007), the anesthetized mice were suspended with their incisors attached to a ∼60° wood support by 3/0 suture. A cold-light source (Dolan-Jenner Industries) with two 25-inch flexible fiber optic arms allowed transillumination of the glottis and vocal cords to deliver LPS, or PBS, into the air spaces of the lung.
Measurement of Lung Extravascular Lung Water (ELW)
As previously described (Su et al.,2007), the lungs were removed, counted in a gamma counter (Packard, Meriden, CT), weighed, and homogenized (after addition of 1 mL distilled water). The blood was collected through right ventricle puncture. The homogenate was weighed and a fraction was centrifuged (12,000 rpm, 8 min) for assay of hemoglobin concentration in the supernatant. Another fraction of homogenate, supernatant, and blood was weighed, and then desiccated in an oven (60°C for 24 hr). We used the following formula to calculate ELW:
- 1Measurement of the water fraction separately in the lung homogenate (WFH), the supernatant of the homogenate (WFS), and the blood (WFB) by:where Wwet is the wet weight, and Wdry is the dry weight.
- 2Calculation of blood volume in the lung:where 1.039 is the density of blood, QH is the weight of lung homogenate (whole lung wet weight + weight of the distilled water), WFH is the water fraction of lung homogenate, HbS is the hemoglobin concentration of supernatant of lung homogenate, WFS is the water fraction in supernatant of lung homogenate, and HbB is the hemoglobin concentration of blood.
- 3Calculation of water volume in the lung:where Ww is the weight of the distilled water added when the lung is homogenized.
- 4Calculation of whole lung dry weight (Qd):
- 5Calculation of the extravascular lung water (ELW):
where QW exp equals water volume of the lung in the experimental group; Qd exp equals dry weight of the lung in the experimental group. The controls were the normal mice with the same age as the experimental group.
Measurement of Lung Vascular Permeability
125I-labeled albumin (Iso-Tex Diagnostics) was given IP 4 hr before the end of the experiment. A gamma counter (Packard 5000 Series) was used to measure the radioactivity of the blood samples and the removed lungs. Lung extravascular plasma equivalents (EVPE; index of lung vascular permeability to protein) were determined as the counts of 125I-albumin in the blood-free lung tissue divided by the counts of 125I-albumin in the plasma (Wiener-Kronish et al.,1991; Folkesson et al.,1995).
Bronchoalveolar Lavage and Cytokine Measurements
Bronchoalveolar lavage (BAL) samples were obtained from mice at 24 hr after the challenge dose of endotoxin. After euthanizing the mice, BAL was performed by placing a 20 gauge catheter into the trachea. The lungs were flushed with 1 mL of cold PBS back and forth three times. The levels of TNF-α and MIP-2 in the BAL were measured by ELISA (R&D Systems). BAL levels of receptor for advanced glycation end products (RAGE) were also measured using Mouse RAGE DuoSet ELISA (R&D Systems).
BAL Total Cell and Neutrophil Counts
BAL total cell count was determined using a Coulter counter (Beckman). Neutrophil differential was determined using cytospin (Thermo Shandon) to make a cell smear. The cells were visualized using Wright-Giemsa staining (Fisher Scientific). A representative portion of the slide was used to count 100 cells and generate a differential of the white blood cells. Neutrophil counts were calculated by multiplying the BAL total cell counts by the percentage of neutrophils.
Alveolar Fluid Clearance Measurements
The in situ method for measuring alveolar fluid clearance (AFC) was used as previously described (Garat et al.,1998). Briefly, 24 hr after receiving the challenge dose of endotoxin, mice were euthanized with an overdose of IP pentobarbital sodium (0.6 mg/g). The abdominal aorta was transected through a small abdominal laparotomy to exsanguinate the mouse. An incision was made in the trachea, and it was cannulated with a 20-gauge trimmed Angiocath plastic needle. Body temperature was maintained at the body temperature before pentobarbital administration using a heating pad and an infrared lamp. An isosmolar 5% albumin solution (400 μL) with 125I-labeled albumin (Iso-Tex Diagnostics) was gently instilled into the lungs. This solution was aspirated and instilled three times for even distribution and mixing. The tracheostomy was connected to continuous positive airway pressure at 5 cm H2O and 100% O2 was delivered. After 1 min, a 50 μL sample was obtained. After 30 min, a syringe was attached to the tracheostomy and any remaining fluid in the airspaces was aspirated. The concentration of 125I-labeled albumin was measured using a gamma counter (Packard 5000 Series). AFC was calculated by measuring the increase in albumin concentration of the instilled albumin solution at 30 min compared with 1 min.
NF-κB p65 in the Lung Nuclear Extract
To study whether LPS priming increases translocation of NF-κB p65 from cytoplasm to the nucleus, mice from the control, unprimed, and primed groups were euthanized 6 hr after IT challenge of LPS or PBS. Both lungs were removed en bloc and homogenized. Lung nuclear extract was obtained using a Nuclear Extract Kit (Active Motif). The levels of NF-κB p65 in the lung nuclear extract were measured by a TransAM NF-κB p65 Transcription Factor Assay Kit (Active Motif).
Lungs from all three groups were excised at 24 hr following the challenge dose of endotoxin. The lungs were fixed with 100% ethanol intratracheally under 20 cm of H2O pressure. After fixation, lungs were embedded in paraffin, cut into 5-μm sections, and stained with hematoxylin and eosin (H&E).
Three sections in each lung from each mouse were cut. Under objective magnification 40×, three fields were randomly chosen to be examined in each section. The degree of microscopic injury was scored based on the following variables: interstitial edema and neutrophil infiltration. The severity of injury was judged by the following criteria: no injury = 0; injury to 25% of the field = 1; injury to 50% of the field = 2; injury to 75% of the field = 3; diffuse injury = 4 (Mrozek et al.,1997; Su et al.,2003; Gupta et al.,2007). The lung injury scoring was performed by a reviewer blinded to the experimental group.
Comparisons between groups were made using ANOVA analysis with Bonferroni adjustment using Stata 9.0 software. A value of P < 0.05 was considered statistically significant. Data are shown as mean ± SD.
Priming With Endotoxin Increased Body Weight Loss and Induced More Hypothermia
Unprimed mice had significantly more body weight loss (Fig. 1A) and hypothermia (Fig. 1B) compared with the control mice at 24 hr. The mice in the primed group had a trend toward increased body weight loss compared with the unprimed group (Fig. 1A) and had significantly more hypothermia (Fig. 1B).
Pulmonary Edema and Endothelial Permeability Were Enhanced in the Mice Primed With Endotoxin
Pulmonary edema, as measured by extravascular lung water (ELW), was increased in the mice that received a single, challenge dose of endotoxin compared with PBS controls (Fig. 2A). When a small priming dose of endotoxin was delivered, the mice had a 47% increase in ELW compared with the unprimed group.
Lung endothelial permeability, measured by EVPE of 125I-albumin, was similar in the unprimed mice and controls (Fig. 2B). However, the EVPE of 125I-albumin was significantly increased in the primed group compared with the unprimed group.
Priming With Endotoxin Impairs Alveolar Fluid Transport but Does Not Augment Lung Epithelial Injury
Alveolar epithelial function was assessed by measuring AFC. AFC was significantly reduced in the primed group compared with the PBS controls (Fig. 3A). However, there was no difference in AFC between the unprimed group and PBS controls.
To test for lung epithelial injury with a biochemical marker, we measured BAL RAGE, a type I alveolar epithelial cell marker that correlates with the severity of lung epithelial injury (Su et al.,2009). BAL levels of RAGE were increased in the unprimed group compared with the PBS control group (Fig. 3B). However, BAL levels of RAGE were not different between the primed and unprimed groups.
Priming With Endotoxin Increased Alveolar Inflammatory Cell Influx
There was an increase in the total number of white blood cells in the BAL of unprimed mice compared with PBS controls (Fig. 4A). In the primed group, the total cell count was significantly increased compared with the unprimed group. Further analysis demonstrated that this was primarily an increase in the absolute neutrophil count in the primed group compared with the unprimed group (Fig. 4B). There were very few neutrophils in the PBS control group.
Endotoxin Priming Worsens Interstitial Lung Injury
In lung histology, both the primed and unprimed groups had increased interstitial edema and injury compared with the control group. Morphologically, more neutrophil infiltration and interstitial edema were found in the primed group compared with the unprimed groups (Fig. 5A–F). Lung injury score for interstitial edema and neutrophil infiltration were also increased in the primed group compared with the unprimed group (Fig. 5G).
Effects of Endotoxin Priming on Activation of NF-κB in the Lung
NF-κB activation was determined by measuring levels of the NF-κB p65 subunit in the lung nuclear extract 6 hr after the challenge dose of endotoxin. At this time point, the levels of NF-κB p65 subunit in the lung nuclear extract were higher in the primed and unprimed groups compared with the control group (Fig. 6A). The levels of NF-κB in the lung nuclear extract were not different in the primed group compared with the unprimed group.
Twenty-four hours after the challenge dose of endotoxin, the levels of BAL MIP-2 were undetectable in the control group (Fig. 6B). There was no difference in the levels of BAL MIP-2 between the primed and unprimed groups. In addition, the levels of BAL TNF-α were not different between the unprimed and primed groups 24 hr following the challenge dose of endotoxin (data not shown).
We have described a novel model of endotoxin-induced ALI using a priming method. The timing of the priming dose of endotoxin was based on previous studies (Wollert et al.,1994; Kabir et al.,2002), as well as clinical relevance. This two-hit model of endotoxin ALI is a more clinically relevant mouse model to study ALI, because it more accurately depicts the combination of insults seen in the human syndrome of ALI/ARDS.
The main findings of these experimental studies in mice can be summarized as follows. Priming with endotoxin led to more systemic signs of illness, including increased body weight loss and hypothermia, compared with a single-hit model. There was also evidence of a greater increase in extravascular lung water and augmented lung endothelial permeability with endotoxin priming. In addition, there was an increase in alveolar inflammation as demonstrated by increased numbers of inflammatory cells, particularly neutrophils, in the BAL of the primed group.
In these studies, the increase in extravascular lung water was primarily due to an increase in lung endothelial permeability in the primed group compared with the unprimed group as demonstrated by the differences in EVPE of 125I-albumin. A modest decrease in AFC in the primed group may have also contributed to the greater increase in excess lung water in the primed group. There was no difference in lung epithelial injury between the unprimed and primed groups as demonstrated by similar levels of BAL RAGE. In this model, the reduced AFC may not be related to enhanced epithelial injury, but perhaps to the increased inflammation in the priming model. Histological examination of the lungs comparing primed and unprimed mice supports this hypothesis of enhanced inflammation in the primed group, as demonstrated by increased interstitial thickness and neutrophil infiltration. Acute inflammation has been demonstrated to reduce AFC independent of epithelial injury (Lee et al.,2007).
The explanation for the increase in neutrophils in the primed group is not explained by our data. Interestingly, the levels of NF-κB p65 in the lung nuclear extract and BAL levels of MIP-2 and TNF-α were no different in the primed group compared with the unprimed group. This suggests that priming does not increase activation of NF-κB during this model of ALI. Other signaling pathways and neutrophil chemokines (e.g., KC) may be involved in the enhanced lung edema and neutrophil recruitment in this priming model of ALI. One potential alternative pathway could be through the MAP kinase family of proteins such as JNK. JNK activates the transcription factor AP-1, which is also a mediator of inflammation (Takeda and Akira,2004).
The priming effect of endotoxin in a double-hit endotoxin injury model has been well described previously and is known as the Shwartzman reaction (Brozna,1990). The Shwartzman reaction is a model of organ injury caused by two consecutive administrations of endotoxin typically leading to the syndrome of disseminated intravascular coagulation. The Shwartzman reaction differs from the phenomenon of endotoxin tolerance, in which repeated administrations of low dose endotoxin leads to protection against endotoxin-induced lethality (West and Heagy,2002). Our model may be consistent with the Shwartzman phenomenon because we found increased lung injury with repeated doses of endotoxin, although we did not assess the effect of our priming model on the coagulation system.
There have been other published reports of endotoxin priming models used by other investigators. Imamura et al. described a similar study using a low IT dose of endotoxin followed by a larger intravenous dose of endotoxin in a rabbit model of ALI (Imamura et al.,1997). They primarily described the cytokine pattern generated and found that TNF-α levels peaked at 0.5 hr following the intravenous dose of endotoxin. Another study described a swine model of endotoxin priming ALI. In their model, the endotoxin delivery was intravenous, but the timing between the two injections was similar to our study (Wollert et al.,1994). They found that priming with endotoxin exacerbated arterial hypoxemia and led to more reproducible and severe lung injury compared with the unprimed group.
There are some limitations to this experimental model. This model is nonlethal and therefore cannot be used to study the most severe forms of ALI/ARDS. However, this model can be used for mechanistic studies of nonlethal ALI/ARDS. It is possible that additional insults could be added to this injury model, for example, mechanical ventilation and hyperoxia, to enhance the level of injury and expand the clinical relevance of the animal model. Another limitation of this experimental model is that we only describe one time point, 24 hr, after the challenge dose of endotoxin, although we believe this is a practical time interval for investigators to use in their research. Lastly, we did not measure the feeding habits of the mice, so we do not know if the weight loss was due to less consumption or increased metabolism.
In summary, priming with a small dose of endotoxin followed by a larger challenge dose of endotoxin induces more systemic illness and increased pulmonary edema, largely due to an increase in lung endothelial permeability and lung inflammation with a modest reduction in AFC. The enhanced injury seen with priming appears to be associated with NF-κB-independent pathways. This model should be useful to other investigators who want to study the effects of priming and clinically relevant insults to the lung in a mouse model of ALI.
The authors would like to thank Hanjing Zhuo for her help in performing the statistical analysis used in this study.
- 1990. Shwartzman reaction. Semin Thromb Hemost 16: 326–332. .
- 1995. Acid aspiration-induced lung injury in rabbits is mediated by interleukin-8-dependent mechanisms. J Clin Invest 96: 107–116. , , , .
- 1998. New in situ mouse model to quantify alveolar epithelial fluid clearance. J Appl Physiol 84: 1763–1767. , , .
- 2007. Intrapulmonary delivery of bone marrow derived mesenchymal stem cells improves survival and attenuates endotoxin-induced acute lung injury in mice. J Immunol 179: 1855–63. , , , , .
- 1997. Involvement of tumor necrosis factor-alpha, interleukin-1 beta, interleukin-8, and interleukin-1 receptor antagonist in acute lung injury caused by local Shwartzman reaction. Pathol Int 47: 16–24. , , , , .
- 2002. Characterization of a murine model of endotoxin-induced acute lung injury. Shock 17: 300–303. , , , , , , , , .
- 2005. One-hit, two-hit . . . is there really any benefit? Clin Exp Immunol 141: 211–214. , .
- 2007. Acute lung injury edema fluid decreases net fluid transport across human alveolar epithelial type II cells. J Biol Chem 282: 24109–24119. , , , , , , .
- 2009. Platelet depletion and aspirin treatment protect mice in a two-event model of transfusion-related acute lung injury. J Clin Invest 119: 3450–3461. , , , , , .
- 2006. Neutrophils and their Fc gamma receptors are essential in a mouse model of transfusion-related acute lung injury. J Clin Invest 116: 1615–1623. , , , , .
- 2008. Animal models of acute lung injury. Am J Physiol Lung Cell Mol Physiol 295:L379–L399. , , .
- 1997. Exogenous surfactant and partial liquid ventilation: physiologic and pathologic effects. Am J Respir Crit Care Med 156: 1058–1065. , , , , , , .
- 2005. Incidence and outcomes of acute lung injury. N Engl J Med 353: 1685–1693. , , , , , , , .
- 2003. Effect of continuous hemofiltration on hemodynamics, lung inflammation and pulmonary edema in a canine model of acute lung injury. Intensive Care Med 29: 2034–2042. , , , , , , , , .
- 2007. Activation of the alpha7 nAChR reduces acid-induced acute lung injury in mice and rats. Am J Respir Cell Mol Biol 37: 186–192. , , , , , , , .
- 2004. Direct visual instillation as a method for efficient delivery of fluid into the distal airspaces of anesthetized mice. Exp Lung Res 30: 479–493. , , , , .
- 2009. Receptor for advanced glycation end-products (RAGE) is an indicator of direct lung injury in models of experimental lung injury. Am J Physiol Lung Cell Mol Physiol 297: L1–L5. , , , .
- 2004. TLR signaling pathways. Semin Immunol 16: 3–9. , .
- 2009. Transfusion-related acute lung injury: current concepts for the clinician. Anesth Analg 108: 770–776. .
- 2000. The acute respiratory distress syndrome. N Engl J Med 342: 1334–1349. , .
- 2002. Endotoxin tolerance: a review. Crit Care Med 30: S64–S73. , .
- 1991. Differential responses of the endothelial and epithelial barriers of the lung in sheep to Escherichia coli endotoxin. J Clin Invest 88: 864–875. , , .
- 1994. Prior exposure to endotoxin exacerbates lipopolysaccharide-induced hypoxemia and alveolitis in anesthetized swine. Shock 2: 362–369. , , , , , , .
- 2000. Molecular mechanisms of NF-kappaB activation induced by bacterial lipopolysaccharide through Toll-like receptors. J Endotoxin Res 6: 453–457. , .