Skip to main content

Type XVIII collagen degradation products in acute lung injury

Abstract

Introduction

In acute lung injury, repair of the damaged alveolar-capillary barrier is an essential part of recovery. Endostatin is a 20 to 28 kDa proteolytic fragment of the basement membrane collagen XVIII, which has been shown to inhibit angiogenesis via action on endothelial cells. We hypothesised that endostatin may have a role in inhibiting lung repair in patients with lung injury. The aims of the study were to determine if endostatin is elevated in the plasma/bronchoalveolar lavage fluid of patients with acute lung injury and ascertain whether the levels reflect the severity of injury and alveolar inflammation, and to assess if endostatin changes occur early after the injurious lung stimuli of one lung ventilation and lipopolysaccharide (LPS) challenge.

Methods

Endostatin was measured by ELISA and western blotting.

Results

Endostatin is elevated within the plasma and bronchoalveolar lavage fluid of patients with acute lung injury. Lavage endostatin reflected the degree of alveolar neutrophilia and the extent of the loss of protein selectivity of the alveolar-capillary barrier. Plasma levels of endostatin correlated with the severity of physiological derangement. Western blotting confirmed elevated type XVIII collagen precursor levels in the plasma and lavage and multiple endostatin-like fragments in the lavage of patients. One lung ventilation and LPS challenge rapidly induce increases in lung endostatin levels.

Conclusions

Endostatin may adversely affect both alveolar barrier endothelial and epithelial cells, so its presence within both the circulation and the lung may have a pathophysiological role in acute lung injury that warrants further evaluation.

Introduction

Acute lung injury (ALI) is characterised by neutrophilic inflammation of the alveolar-capillary barrier. ALI has multiple aetiologies, but appears to follow a uniform pattern of injury at a cellular level. Extensive damage to the alveolar-capillary barrier leads to the influx of a protein-rich oedema fluid and accompanying inflammatory cells into the alveoli. A complex cascade of both inflammatory and anti-inflammatory cytokines is triggered and inflammatory cells, including neutrophils and monocytes, are recruited to the alveoli. Studying these processes early in the course of the disease can be challenging because most insults causing lung injury are not predictably timed.

Two human models of lung injury allow assessment of the early phases of ALI. One lung ventilation (OLV) during oesophagectomy is associated with a significant post-operative risk of ALI with proposed causative mechanisms including the ischaemic/reperfusion insult experienced by the collapsed lung, oxidative stress injury and barotrauma to the ventilated lung [1]. Systemic levels of inflammatory cytokines have been shown to relate to the duration of OLV and to be reduced by temporary ventilation of the collapsed lung [2]. In addition, recently there have been several studies looking at lipopolysaccharide (LPS) challenge in human volunteers. This model induces a significant neutrophilia in the bronchoalveolar lavage (BAL) with disruption of the alveolar-capillary barrier and allows early pathophysiological changes within defined time limits to be assessed.

The neutrophilic inflammation of the alveolar capillary barrier in ALI and the models of OLV or LPS challenge result in the release of proteases including collagenases [3]. The balance between collagen formation and degradation is a complex and dynamic process within the lung of patients with ALI [4]. BAL studies suggest changes in collagen production and degradation may promote collagen deposition within the lung, even at the onset of lung injury [4, 5].

There are at least 27 different species of collagen. Types I and III predominate within both healthy and fibrotic lung [6]. Perivascular tissue contains type XVIII collagen which is expressed as three variable polypeptide forms (SHORT, MIDDLE and LONG/frizzled) [7] (Figure 1). Endostatin is a 20 kDa proteolytic fragment of collagen XVIII. Recombinant endostatin has been shown to inhibit tumour growth and metastasis in animal models [8]. On the cellular level, endostatin specifically blocks growth factor-induced proliferation and migration of endothelial cells. The latter is proposed to involve integrin binding and subsequent disruption of the cell-matrix interaction either via Src tyrosine kinase/Rho pathway or mitogen activated protein kinase (MAPK)/p38 pathway [9–11]. Endostatin induces endothelial cell apoptosis in microgram doses [12], inhibits vascular endothelial growth factor (VEGF)-mediated signalling due to a direct interaction with VEGF receptor-2 [13], and inhibits cyclin D1 [14] and Wnt signalling [15]. The main focus of research into endostatin has been its anti-tumour effects. Recently, however, elevated levels have been found in the plasma of patients with preeclampsia, a condition also associated with pan-endothelial damage [16]. There is little information about the effects of endostatin on epithelial cells; however, it was shown to inhibit squamous cell carcinoma migration and invasion in vitro [17, 18].

Figure 1
figure 1

Schematic representation of the human collagen XVIII variants, termed as SHORT, MIDDLE and LONG/FZ. Collagenous sequences are shown in white. Non-collagenous (NC) amino terminal sequences common to all variants are shown in black. Non-collagenous amino terminal sequences common to the two long variants are shown in grey. A non-collagenous amino terminal sequence specific to the LONG/FZ variant is marked with waves. The amino acid lengths (aa) and the predicted molecular masses (kDa) of these sequences are given, as are those of the full length type XVIII collagen molecules. Signal sequences are indicated by vertical and horizontal hatching. The epitopes of the antibodies used in this work are shown by black horizontal lines. The molecular masses of some endostatin-containing carboxy-terminal fragments are also indicated.

Resolution of ALI is dependent on the successful repair of a confluent barrier of alveolar epithelial and endothelial cells [19]. This process requires cell division and migration. Molecules that effect cell proliferation, migration and repair are therefore potential therapeutic agents in ALI [20]. As collagen degradation products are elevated in the lungs of patients with ALI [4], we hypothesised that endostatin may have a role in inhibiting lung repair in patients with ALI. The aim of this study, therefore, was to test the hypothesis that endostatin is elevated in the plasma and BAL of patients with ALI and to determine whether endostatin levels reflect the severity of injury and alveolar inflammation. In addition, we assessed changes in endostatin in OLV and LPS challenge models of early lung injury.

Materials and methods

Subjects

All patients enrolled in the study gave written informed consent themselves at the time of enrollment, or it was given by the consultant in charge of the intensive care unit (ICU) (not part of the research team) as their legal guardian if they were unable to give consent because of sedation and ventilation. In addition, in all cases, patients' next of kin gave informed signed assent to inclusion. In the UK. patients' relatives cannot consent for incapacitated relatives. In addition, those patients who recovered mental competency during their inpatient stay retrospectively provided signed consent for their inclusion in the study. This study was fully approved by the local National Health Service trusts and ethical review committee.

Patients were studied within 48 hours of admission to the ICU of Birmingham Heartlands and University Hospital, Birmingham, UK, between 2006 and 2008. 38 patients (22 male, mean age 62 years, standard deviation (SD) 16 years, 12 with direct and 21 with remote lung injury) were identified as having ALI according to the American-European consensus statement [21]. Patients were ventilated using pressure-controlled ventilation. Bronchoscopy and BAL was performed in all patients immediately following inclusion and, when possible, four days later (n = 22). The median time between patients satisfying the criteria for ALI and initial bronchoscopy was about six hours (range = 3 to 36) Of the patients who did not undergo repeat bronchoscopy, six had died, five had been extubated and five had contraindications. Ten ventilated patients with identifiable risk factors for ALI but who did not satisfy the consensus criteria for ALI were recruited as controls (6 male, mean age 59 years, SD 15 years). Ten non-smokers underwent bronchoscopy as normal controls.

Patient demographic characteristics were recorded at baseline. The acute physiology and chronic health evaluation II (APACHE II) and simplified acute physiology score II (SAPS II) were recorded as global markers of disease severity. The Murray lung injury score, partial pressure of arterial oxgen (PaO2): fraction of inspired oxygen (FiO2) ratio was recorded on the day of bronchoscopy. A summary of patient physiological severity is shown in Table 1.

Table 1 Characteristics of acute lung injury and septic at-risk patients

Post-one lung ventilation patients

Ten patients (9 males, 1 female, age 63 years, mean forced expiratory volume (FEV) 1 94% predicted, mean 25 pack-year smoking history) undergoing elective oesophagectomy for oesophageal carcinoma underwent BAL of the collapsed lung at the end of the operation, which as such potentially suffers from ischaemia-reperfusion injury. Ventilation was performed with pressure control for eight patients and two patients with volume control. Mean peak airway pressures were 25.8 cm, standard error of the mean (SEM) 0.88 cm, and a mean tidal volume of 430 ml was achieved. Mean duration of OLV was 189 minutes, SEM 22.29 minutes. The preoperative PaO2:FiO2 ratio was a mean of 50.210, SEM 3.45, and immediate post-operative PaO2:FiO2 ratio was a mean of 40.2, SD 14.46, SEM 4.57, and average intraoperative FiO2 during OLV was 67%. None of these patients went on to develop early lung injury.

Inhaled lipopolysaccharide-induced model of lung injury

We include data from a human LPS challenge study assessing the anti-inflammatory effects of simvastatin versus placebo in healthy volunteers. Data are presented from 10 patients treated with placebo (five males, mean age 26 years) who underwent bronchoscopy six hours post-inhalation of LPS. Full details of the patient characteristics and lavage fluid abnormalities are available in another paper [22].

Bronchoscopy and bronchoalveolar lavage

Bronchoscopy was performed by instilling three 50 ml aliquots into the right middle lobe (median yield in ALI patients = 65 ml, IQR = 35 to 80 ml). Bronchoalveolar lavage fluid (BALF) was processed as described previously [23]. Into lithium heparin tubes (Becton Dickinson, London, UK), 15 ml of blood was collected simultaneously and stored on ice until processing.

Lipopolysaccharide challenge

LPS (Escherichia coli serotype O26:B6; Sigma Chemicals, Poole, Dorset, UK) was dissolved in endotoxin-free sterile 0.9% saline and inhaled via an automatic inhalation–synchronised dosimeter nebuliser (Spira Electra 2, Hameenlinna, Finland), which delivers particles of a mass median aerodynamic diameter of 10 mm, as described previously. The dosimeter produces a calibrated aerosol of 8 μl at each slow inhalation starting from functional residual capacity to total lung capacity. Each subject performed five successive inhalations of the LPS solution (1.25 mg/ml) through the mouthpiece with a nose clip in place. The total dose of inhaled LPS was 50 μg. BAL was performed six hours after LPS inhalation according to standard guidelines by a single researcher (DM).

Urea and protein determination

Urea concentration was determined using a commercially available urea kit (Sigma diagnostics, Poole, UK). The epithelial lining fluid (ELF) levels of endostatin were estimated using the following formula: ELF endostatin = BAL urea (mmol/L) × BAL endostatin/plasma urea concentration (mmol/L), as described previously [24]. BALF protein was measured using the Lowry method [25]. The protein permeability index was calculated as plasma protein/BAL protein [26].

ELISA

Endostatin and IL-8 were measured by using an ELISA kit (R&D systems, Abingdon, UK) according to manufacturer's instructions. CV for intra-assay variability in BALF of endostatin is 3.6% and inter-assay variability was 8%. Spiking ALI BALF with recombinant endostatin standard (2.5 ng/ml) has a 96% recovery rate.

Western blotting of BALF and plasma samples

Immunoprecipitation and western blotting of plasma and BALF was performed using methods described previously [27, 28]. The specificities of the two polyclonal human collagen XVIII antibodies, anti-ALL huXVIII (QH48.18) and anti-LONG huXVIII (QH1415.7), have been verified earlier using competitive blot with the peptides that were used to generate these antibodies [7, 27, 29]. The following in house-developed antibodies were used in these studies in order to detect the precursor forms of type XVIII collagen and any degradation products: Anti-LONG huXVIII is a polyclonal anti-LONG human type XVIII collagen antibody which targets the N-terminal end of the two longer isoforms of XVIII collagen [30]. Anti-ALL huXVIII is a polyclonal human type XVIII antibody which targets the N-terminal end common to all three collagen XVIII isoforms [7]. A polyclonal anti-endostatin antibody known to detect proteolytically cleaved endostatin fragments [30].

Western blotting of BALF samples

A 25 μl sample of BALF was loaded onto polyacrylamide gel and the proteins were separated on a 7 to 12% SDS-PAGE under reducing conditions, electrotransferred to a nitrocellulose membrane (Protran; Schleicher & Schuell, Dassel, Germany) and probed with rabbit polyclonal antibodies against endostatin (HES.6) [28] and human collagen XVIII (anti-all huXVIII QH4818 and anti-long huXVIII, QH1415.7) [7], all used at a concentration 1 μg/ml in 5% fat-free milk powder in 1 × PBS, followed by a horseradish peroxidase-conjugated goat anti-rabbit antibody (Bio-Rad, Abingdon, UK). After washing the membrane extensively with 1 × PBS/0.1% Tween, the proteins that were reactive to collagen XVIII and endostatin antibodies were visualised with ECL western blotting detection reagents (Amersham Pharmacia Biotech, Amersham, UK). For binding sites of antibodies to type XVIII collagen see Figure 1.

Immunoprecipitation and western blotting of plasma samples

Collagen XVIII was immunoprecipitated from human plasma using a monoclonal anti-ALL type XVIII antibody, DB144-N2 bound to anti-mouse IgG-coated magnetic beads (Dynabeads M-280; Dynal, Malmo, Sweden), as described elsewhere [31]. The bound polypeptides were separated on a 7% SDS-PAGE under reducing conditions, electrotransferred to a nitrocellulose membrane and probed with rabbit polyclonal antibodies anti-all huXVIII (QH48.18) or anti-long huXVIII (QH1415.7) [7], both at a concentration of 1 μg/ml in 5% fat-free milk powder in 1 × PBS, followed by a horseradish peroxidase-conjugated goat anti-rabbit antibody. After washing the membrane extensively, the proteins reactive to collagen XVIII antibodies were visualised with ECL western blotting detection reagents. For binding sites of antibodies to type XVIII collagen see Figure 1.

Statistical methods

The Ryan-Joiner normality test was used to test the distribution of the data. Non-parametric Mann-Whitney U tests were used for data that was not normally distributed (ELISA endostatin ALI plasma data and BALF data) and are quoted as medians (interquartile ranges (IQR)). Normally distributed data (optical density data for blots) are quoted as mean (SEM) and compared using analysis of variance (ANOVA). Statistical significance assumed at P < 0.05.

Results

Plasma levels of endostatin are elevated in patients with acute lung injury

Median plasma levels of endostatin were elevated in patients at the onset of ALI (182 ng/ml, IQR = 111 to 244) compared with normal controls (median 96.3 ng/ml, P = 0.008) and patients at risk from ALI (102 ng/ml, P = 0.0094). On day four, plasma levels in ALI patients remained elevated (182.8 ng/ml, IQR = 97.8 to 284.6; Figure 2).

Figure 2
figure 2

Plasma endostatin levels in patients groups. Endostatin was measured by ELISA. Endostatin is significantly elevated in patients with acute lung injury compared with normal and at-risk controls. ALI = acute lung injury.

There were significant linear associations between plasma endostatin at the onset of ALI and the global severity of illness markers APACHE II (r = 0.37, P = 0.038) and SAPS II (r = 0.44, P = 0.007) but not with lung injury score (r = -0.3, P = 0.9) (data not shown). Plasma endostatin levels at the onset of ALI or at day four did not predict whether patients subsequently survived or died.

BALF endostatin is significantly elevated compared with normal and at-risk patients

ALI patients had greater endostatin levels in BALF on both day 0 (median ALI 2.6 ng/ml, IQR = 0.24 to 0.64, P = 0.001) and day 4 (median ALI 1.36 ng/ml, IQR = 0.72 to 2.63, P = 0.01) than healthy individuals (0.08 ng/ml). ALI day 0 endostatin levels (2.6 ng/ml) were higher than those at risk of ALI (0.6 ng/ml, IQR = 0.3 to 1.4, P = 0.0017; Figure 3).

Figure 3
figure 3

BALF endostatin in different patient groups. Bronchoalveolar lavage fluid (BALF) endostatin is significantly elevated in the BALF of patients with acute lung injury compared with normal and at-risk controls. Lavage was repeated where possible again at day 4 in 23 patients. Levels fell significantly from day 0 to day 4 but remained elevated compared with normal and at-risk controls. ALI = acute lung injury.

Endostatin levels in ALI BALF fell from day 0 to day 4 (day 0 median 2.6 ng/ml, IQR = 1.4 to 5.1; vs day 4 median 1.36 ng/ml, IQR = 0.76 to 2.9; P = 0.02). There was no difference between BALF levels at day 0 or day 4 between patients who died or survived, or those with direct or remote lung injury. BALF endostatin did not correlate with APACHE II, SAPS II or lung injury score at day 0 or day 4.

ALI BALF endostatin correlates with BAL neutrophilia and protein permeability index

There was also a significant relationship between day 0 BALF endostatin levels and protein permeability index (r = 0.441, P = 0.013). In addition, day 0 BALF endostatin levels correlated with the degree of BAL neutrophilia (r = 0.417, P = 0.034) but not with total cell count or BALF IL-8. At day 4 endostatin also correlated strongly with protein permeability index (r = 0.723, P = 0.002) and total neutrophil counts (r = 0.76, P = 0.007) (data not shown).

Plasma endostatin is higher than ELF endostatin

Plasma levels of endostatin were significantly higher than calculated ELF levels in the lung (median 182 ng/ml vs ELF 119 ng/ml, 95% confidence interval (CI) diff = 18 to 169; P = 0.016). When ELF levels were lower than plasma levels there was a significant correlation (r = 0.47, P = 0.001) between plasma and ELF endostatin at the onset of ALI (data not shown). However, in some patients, calculated ELF levels were actually higher than plasma suggesting intra-alveolar generation of endostatin. Interestingly, at day 4 ELF endostatin correlated with lung injury score (r = 0.499, P = 0.049) but not in ALI patients at day 0.

OLV and LPS challenge induce a BAL neutrophilia and increase PPI

OLV and LPS challenge induced a BAL neutrophilia and an increase in protein permeability index (PPI) compared with normal controls (PPI in OLV: median = 0.00636, IQR = 0.00389 to 0.01188; P = 0.013 vs normal; PPI in LPS: median = 0.0025, IQR = 0.0013 to 0.003, P = 0.011 versus normal; normal PPI = 0.0014, IQR = 0.0008 to 0.0024). Percentage of neutrophils in BAL rose from 1.1% (IQR = 0.0 to 2.65) in normal BALF to median 11% (IQR = 2 to 60.5) in OLV (P = 0.011), and 41% (IQR = 24 to 57) in LPS challenge as described previously [22].

OLV and LPS challenge increases BALF endostatin

Inhalation challenge with LPS increased BALF endostatin from a median of 0.08 ng/ml in normal individuals to 0.385 ng/ml (IQR = 0.285 to 0.448, P = 0.001) in LPS-challenged patients. OLV increased BAL endostatin (median compared with normal and LPs challenge; Figure 4). BALF endostatin correlated with PPI in the LPS challenge (r = 0.63, P = 0.006) but not in OLV patients. In contrast, plasma levels of endostatin were not altered by LPS challenge or OLV (data not shown).

Figure 4
figure 4

BALF endostatin in the different at-risk groups compared with normal individuals. BALF = bronchoalveolar lavage fluid; LPS = lipopolysaccharide; OLV = one lung ventilation.

Type XVIII collagen precursors are elevated in both the plasma and BALF of patients with ALI compared with normal controls

In exploratory experiments performed in five consecutive non-selected patients, plasma and BALF from patients with ALI demonstrated increased density and number of fragments of the circulating precursor forms of type XVIII collagen compared with normal controls. Whole lane densitometric analysis of the western blot by anti-LONG antibody confirmed increased plasma precursors in ALI (mean optical density (OD) = 198, SE 38) compared with normal plasma (mean OD 79.8, SE 7.9, 95% CI = 11.6 to 226, P = 0.037) (representative gels are shown in Figure 5). Western blotting of the precipitated proteins with three different antibodies to human collagen XVIII (anti-ALL, anti-LONG and anti-endostatin) revealed a main 120 kDa band, which is presumably derived from the middle variant of collagen, containing the N-terminus and the collagenous central part of the molecule but lacking the endostatin domain (Figure 5 and data not shown). Based on the current knowledge of the structure and properties of collagen XVIII, the diffuse bands over 170 kDa in the plasma from some ALI patients are interpreted to represent the long, potentially glycosylated, forms of collagen XVIII.

Figure 5
figure 5

Immunoprecipitation with anti-ALL antibody and detection with anti-LONG huXVIII antibody. Elevated type XVIII collagen precursors in the plasma of ALI patients compared with normal controls is shown. ALI = acute lung injury.

There was also a clear increase in collagen XVIII fragments in BALF detected by the anti-ALL antibody revealing a mean adjusted whole lane OD of 94.9 (SE 9.3) in ALI compared with 41.21 (SE 4.8) (95% CI = 26.6 to 80.6; P = 0.004) in normal controls (Figure 6). The anti-ALL antibody does not bind the carboxy end of type XVIII collagen, so this suggests that type XVIII collagen precursors are present within the BALF of patients with ALI. Similarly, using the anti-LONG antibody, there was evidence of increased BALF levels of the long forms of collagen XVIII (ALI OD = 54.74 (SE 2.1), normal OD = 40.39, 95% CI = 6.8 to 21.7, P = 0.004) (data not shown). The main bands of 55 kDa, also detectable with anti-LONG antibody, and 72 kDa were interpreted to represent proteolytic N-terminal fragments from the middle form and the short form of collagen XVIII, respectively (Figure 5 and data not shown).

Figure 6
figure 6

Western blot by anti-ALL huXVIII antibody. It shows elevated type XVIII collagen precursors in the bronchoalveolar lavage fluid (BALF) of patients with acute lung injury compared with normal controls. ALI = acute lung injury.

BALF from ALI patients contain multiple endostatin-like fragments

Western blotting of BALF with the HES.6 antibody that recognises the carboxyl (endostatin) end of type XVIII collagen revealed increased levels of C-terminal endostatin-containing fragments of collagen XVIII are present within the BALF of ALI patients with a fragmentation pattern that was different to normal controls.

Control samples contained a strong 55 kDa band extending to the central collagenous domain, full-length collagen XVIII precursors between 130 and 170 kDa, and some samples also showed a clear 20 kDa endostatin band. In addition to these bands, ALI samples showed a pattern of endostatin-containing fragments varying in size from 22 to 50 kDa. Histidine-labelled recombinant endostatin was used as a positive control (Mw 22 kDa). Whole lane densitometric analysis showed that significantly elevated amounts of type XVII carboxy terminal endostatin-like fragments were present in ALI samples (mean OD = 54, SE 3) compared with normal BALF (mean OD = 43.9, SE 3.87, 95% CI = 2 to 19, P = 0.02) (representative gel shown in Figure 7).

Figure 7
figure 7

Western blot by anti-endostatin antibody HES.6. It shows endostatin-like degradation products in bronchoalveolar lavage fluid (BALF) of patients with acute lung injury. Positive control was a recombinant human endostatin carrying a histidine tag (molecular weight of 22 kda). ALI = acute lung injury.

Discussion

This study has demonstrated that endostatin is elevated within the plasma and BALF of patients with ALI. Endostatin levels in ALI BALF reflected the degree of neutrophilia and the extent of the loss of protein selectivity of the alveolar-capillary barrier. Plasma levels of endostatin at the onset of ALI were associated with the severity of physiological derangement. Western blotting confirmed elevated type XVIII collagen precursor levels and multiple amino- and carboxy-terminal fragments in plasma and BALF. Increases in endostatin occur early after OLV and LPS challenge in human volunteers but this is compartmentalised to the lung.

Endostatin was detectable in the plasma of all normal subjects studied suggesting a physiological role in the homeostasis of angiogenesis. Elevated endostatin levels have previously been reported in the plasma of patients with preeclampsia, a condition associated with pan-endothelial damage [16]. Similarly, in this study, median plasma level of endostatin in ALI patients were elevated in comparison to both those at risk and normal controls. Plasma levels of endostatin correlated with the severity of the systemic physiological insult rather than lung injury per se. The observed relationship between BALF endostatin and protein permeability index, and the apparent plasma endostatin: ELF endostatin gradient in most of our patients further suggest that alveolar levels of endostatin are raised at least in part because of leakage from the microvasculature into the alveolus at sites of inflammation of the alveolar-capillary membrane. However, in some patients, calculated ELF levels were higher than plasma supporting a role for alveolar generation of endostatin from the type XVIII collagen precursors that we have shown are present.

Endostatin was also elevated in the BALF but not plasma of patients with sepsis who were at risk of ALI. In order to study the time course of changes in endostatin, we evaluated changes in response to two injurious stimuli: OLV and LPS challenge. Both these stimuli induce BAL neutrophilia and increased the PPI in the volunteers/patients. Endostatin was elevated in the BALF but not plasma of the patients after an average of 2.5 hours of OLV and after six hours in the LPS-challenged patients. Thus, endostatin release appears to be compartmentalised to the lung early after such localised injurious stimuli. The relationships of BALF endostatin with PPI in these in vivo models of the early response to ALI and in established ALI suggest that endostatin generation might be useful as a measure of alveolar capillary damage and support a potential pathophysiological role.

Endostatin is derived from collagen XVIII which is a major proteoglycan of both endothelial and epithelial basement membrane zones. Specialised capillaries, such as lung alveoli, express type XVIII collagen in their basement membranes. Three forms of collagen XVIII precursor, designated the SHORT, the MIDDLE and the LONG/frizzled variants, are differentially expressed in normal human tissues. The SHORT form predominates in vascular basement membranes [7], whereas the MIDDLE and the LONG forms are abundantly expressed in hepatocytes and are circulating plasma proteins [31, 32]. Furthermore, proteolytic degradation of type XVIII collagen can generate multiple carboxy-terminal fragments of precursor collagen XVIII [33, 34]. Clearly, therefore, elevated endostatin levels could arise because of cleavage of elevated circulating LONG forms and/or cleavage of the SHORT form at sites of inflammation within the alveolar-capillary membrane.

Our immunoprecipitation and western blot results suggest that degradation products of the LONG type XVIII collagen are elevated in both the plasma and BALF of patients with ALI. Proteolytic cleavage of the carboxy end of collagen XVIII has been demonstrated for many enzymes including elastases, cathepsins and matrix metalloproteinases (MMP), which have been implicated in the pathogenesis of ALI [28, 34, 35]. Western blotting of ALI BALF with antibody against the carboxy-terminal of collagen XVIII demonstrated multiple endostatin-like fragments similar in size to those produced by MMP degradation. These fragments may be important because they are known to be bioactive and inhibit β-fibroblast growth factor-induced endothelial cell proliferation and migration [28].

What are the implications of our findings for alveolar capillary repair in ALI? Previous studies using the animal corneal micropocket assay have demonstrated that BALF has a strong angiogenic potential that is related to elevated CXC chemokine levels [36]. In contrast, when looking at human primary lung microvascular endothelial cells, ALI BALF caused cell death in a TNF- and oncostatin-dependent manner [37]. Given the known anti-endothelial actions of endostatin, the elevated levels of endostatin seen in our patients may play a role in such endothelial toxicity.

In addition to effects on endothelial cells, we have recently reported that endostatin inhibits the proliferation and in vitro wound repair responses of both distal small airway epithelial cells, and primary human type II epithelial cells [38]. Thus elevated levels of endostatin within the lung may also play a role in aberrant epithelial repair mechanisms in ALI. The observed relationships in this study between endostatin levels, the degree of neutrophilic inflammation and physiological severity suggest the multiple cellular effects of endostatin within the lung might be of clinical importance.

This study has several limitations. Firstly, despite several attempts, we were unable to detect plasma endostatin fragments by immunoprecipitation and western blotting with the HES.6 antibody, which appears unsuitable for plasma endostatin estimation.

Secondly, there were a number of drop outs in our sequential assessments, for clinical reasons (extubation, death or contraindication to bronchoscopy). We do not believe the results have been biased by this because there were no baseline differences in endostatin between those re-studied at day 4 and those not. Thirdly, it would be interesting to observe the effect of neutralising endostatin bioactivity in BALF on endothelial cell viability. Unfortunately, no effective specific inhibitor for the bioactivity of endostatin is currently available as endostatin has a wide variety of effects on endothelial cells influencing nearly 12% of the genome via a variety of mechanisms [39]. Clearly in order to firmly establish the pathophysiological importance of endostatin release in the development of lung injury further study in animal knock-out models would be required.

Conclusions

To our knowledge, this is the first report of the presence of endostatin in plasma and BALF of humans with ALI. It is clear that ALI patients have persistently elevated alveolar endostatin levels during the early course of the disease. Our models of the early onset of lung injury suggest these changes occur very early in the injurious process. As endostatin may adversely affect both alveolar-barrier endothelial and epithelial cells, its presence within both the circulation and the lung may have a pathophysiological role in ALI that warrants further evaluation.

Key messages

  • Endostatin is elevated within the plasma and BALF of patients with ALI.

  • Endostatin levels in ALI BALF reflected the degree of neutrophilia and the extent of the loss of protein selectivity of the alveolar-capillary barrier.

  • Plasma levels of endostatin at the onset of ALI were associated with the severity of physiological derangement.

  • Western blotting confirmed elevated type XVIII collagen precursor levels and multiple amino- and carboxy-terminal fragments in plasma and BALF.

  • Increases in endostatin occur early after OLV and LPS challenge in human volunteers but this is compartmentalised to the lung.

Abbreviations

ANOVA:

analysis of variance

ALI:

acute lung injury

APACHE II:

acute physiology and chronic health evaluation II

BAL:

bronchoalveolar lavage

BALF:

bronchoalveolar lavage fluid

ELF:

epithelial lining fluid

ELISA:

enzyme-linked immunosorbent assay

FEV:

forced expiratory volume

FiO2:

fraction of inspired oxygen

ICU:

intensive care unit

Ig:

immunoglobulin

IL:

interleukin

IQR:

interquartile range

LPS:

lipopolysaccharide

MAPK:

mitogen activated protein kinase

MMP:

matrix metalloproteinases

OD:

optical density

OLV:

one lung ventilation

PaO2:

partial pressureof arterial oxygen

PBS:

phosphate-buffered saline

SAPS II:

simplified acute physiology score II

SD:

standard deviation

SEM:

standard error of the mean

TNF:

tumour necrosis factor

VEGF:

vascular endothelial growth factor.

References

  1. Baudouin S: Lung injury after thoracotomy. Br J Anaesth 2003, 91: 132-142. 10.1093/bja/aeg083

    Article  CAS  PubMed  Google Scholar 

  2. Ojima H, Kuwano H, Kato H, Miyazaki T, Nakajima M, Sohda M, Tsukada K: Relationship between cytokine response and temporary ventilation during one-lung ventilation in esophagectomy. Hepatogastroenterology 2007, 54: 111-115.

    CAS  PubMed  Google Scholar 

  3. Weiland JE, Davis WB, Holter JF, Mohammed JR, Dorinsky PM, Gadek JE: Lung neutrophils in the adult respiratory distress syndrome. Clinical and pathophysiologic significance. Am Rev Respir Dis 1986, 133: 218-225.

    CAS  PubMed  Google Scholar 

  4. Armstrong L, Thickett DR, Mansell JP, Ionescu M, Hoyle E, Billinghurst RC, Poole AR, Millar AB: Changes in collagen turnover in early acute respiratory distress syndrome. Am J Respir Crit Care Med 1999, 160: 1910-1915.

    Article  CAS  PubMed  Google Scholar 

  5. Marshall RP, Bellingan G, Webb S, Puddicombe A, Goldsack N, McAnulty RJ, Laurent GJ: Fibroproliferation occurs early in the acute respiratory distress syndrome and impacts on outcome. Am J Respir Crit Care Med 2000, 162: 1783-1788.

    Article  CAS  PubMed  Google Scholar 

  6. Thickett DR, Poole AR, Millar AB: The balance between collagen synthesis and degradation in diffuse lung disease. Sarcoidosis Vasc Diffuse Lung Dis 2001, 18: 27-33.

    CAS  PubMed  Google Scholar 

  7. Saarela J, Rehn M, Oikarinen A, Autio-Harmainen H, Pihlajaniemi T: The short and long forms of type XVIII collagen show clear tissue specificities in their expression and location in basement membrane zones in humans. Am J Pathol 1998, 153: 611-626.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  8. O'Reilly MS, Boehm T, Shing Y, Fukai N, Vasios G, Lane WS, Flynn E, Birkhead JR, Olsen BR, Folkman J: Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell 1997, 88: 277-285. 10.1016/S0092-8674(00)81848-6

    Article  PubMed  Google Scholar 

  9. Wickstrom SA, Veikkola T, Rehn M, Pihlajaniemi T, Alitalo K, Keski-Oja J: Endostatin-induced modulation of plasminogen activation with concomitant loss of focal adhesions and actin stress fibers in cultured human endothelial cells. Cancer Res 2001, 61: 6511-6516.

    CAS  PubMed  Google Scholar 

  10. Wickstrom SA, Alitalo K, Keski-Oja J: Endostatin associates with integrin alpha5beta1 and caveolin-1, and activates Src via a tyrosyl phosphatase-dependent pathway in human endothelial cells. Cancer Res 2002, 62: 5580-5589.

    CAS  PubMed  Google Scholar 

  11. Sudhakar A, Sugimoto H, Yang C, Lively J, Zeisberg M, Kalluri R: Human tumstatin and human endostatin exhibit distinct antiangiogenic activities mediated by alpha v beta 3 and alpha 5 beta 1 integrins. Proc Natl Acad Sci USA 2003, 100: 4766-4771. 10.1073/pnas.0730882100

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  12. Dhanabal M, Ramchandran R, Waterman MJ, Lu H, Knebelmann B, Segal M, Sukhatme VP: Endostatin induces endothelial cell apoptosis. J Biol Chem 1999, 274: 11721-11726. 10.1074/jbc.274.17.11721

    Article  CAS  PubMed  Google Scholar 

  13. Kim YM, Hwang S, Pyun BJ, Kim TY, Lee ST, Gho YS, Kwon YG: Endostatin blocks vascular endothelial growth factor-mediated signaling via direct interaction with KDR/Flk-1. J Biol Chem 2002, 277: 27872-27879. 10.1074/jbc.M202771200

    Article  CAS  PubMed  Google Scholar 

  14. Hanai J, Dhanabal M, Karumanchi SA, Albanese C, Waterman M, Chan B, Ramchandran R, Pestell R, Sukhatme VP: Endostatin causes G1 arrest of endothelial cells through inhibition of cyclin D1. J Biol Chem 2002, 277: 16464-16469. 10.1074/jbc.M112274200

    Article  CAS  PubMed  Google Scholar 

  15. Hanai J, Gloy J, Karumanchi SA, Kale S, Tang J, Hu G, Chan B, Ramchandran R, Jha V, Sukhatme VP, Sokol S: Endostatin is a potential inhibitor of Wnt signaling. J Cell Biol 2002, 158: 529-539. 10.1083/jcb.200203064

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  16. Hirtenlehner K, Pollheimer J, Lichtenberger C, Wolschek MF, Zeisler H, Husslein P, Knofler M: Elevated serum concentrations of the angiogenesis inhibitor endostatin in preeclamptic women. J Soc Gynecol Investig 2003, 10: 412-417. 10.1016/S1071-5576(03)00142-4

    Article  CAS  PubMed  Google Scholar 

  17. Nyberg P, Heikkila P, Sorsa T, Luostarinen J, Heljasvaara R, Stenman UH, Pihlajaniemi T, Salo T: Endostatin inhibits human tongue carcinoma cell invasion and intravasation and blocks the activation of matrix metalloprotease-2, -9, and -13. J Biol Chem 2003, 278: 22404-22411. 10.1074/jbc.M210325200

    Article  CAS  PubMed  Google Scholar 

  18. Wilson RF, Morse MA, Pei P, Renner RJ, Schuller DE, Robertson FM, Mallery SR: Endostatin inhibits migration and invasion of head and neck squamous cell carcinoma cells. Anticancer Res 2003, 23: 1289-1295.

    CAS  PubMed  Google Scholar 

  19. Berthiaume Y, Lesur O, Dagenais A: Treatment of adult respiratory distress syndrome: plea for rescue therapy of the alveolar epithelium. Thorax 1999, 54: 150-160.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  20. Lesur O, Berthiaume Y, Blaise G, Damas P, Deland E, Guimond JG, Michel RP: Acute respiratory distress syndrome: 30 years later. Can Respir J 1999, 6: 71-86.

    CAS  PubMed  Google Scholar 

  21. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, Lamy M, Legall JR, Morris A, Spragg R: The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994, 149: 818-824.

    Article  CAS  PubMed  Google Scholar 

  22. Shyamsundar M, McKeown ST, O'Kane CM, Craig TR, Brown V, Thickett DR, Matthay MA, Taggart C, Backman JT, Elborn JS, McAuley DF: Simvastatin decreases lipopolysaccharide induced pulmonary inflammation in healthy volunteers. Am J Respir Crit Care Med 2009, in press.

    Google Scholar 

  23. Perkins GD, Roberts J, McAuley DF, Armstrong L, Millar A, Gao F, Thickett DR: Regulation of vascular endothelial growth factor bioactivity in patients with acute lung injury. Thorax 2005, 60: 153-158. 10.1136/thx.2004.027912

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  24. Thickett DR, Armstrong L, Millar AB: A role for vascular endothelial growth factor in acute and resolving lung injury. Am J Respir Crit Care Med 2002, 166: 1332-1337. 10.1164/rccm.2105057

    Article  PubMed  Google Scholar 

  25. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the Folin phenol reagent. J Biol Chem 1951, 193: 265-275.

    CAS  PubMed  Google Scholar 

  26. Holter JF, Weiland JE, Pacht ER, Gadek JE, Davis WB: Protein permeability in the adult respiratory distress syndrome. Loss of size selectivity of the alveolar epithelium. J Clin Invest 1986, 78: 1513-1522. 10.1172/JCI112743

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  27. Musso O, Rehn M, Saarela J, Theret N, Lietard J, Hintikka , Lotrian D, Campion JP, Pihlajaniemi T, Clement B: Collagen XVIII is localized in sinusoids and basement membrane zones and expressed by hepatocytes and activated stellate cells in fibrotic human liver. Hepatology 1998, 28: 98-107. 10.1002/hep.510280115

    Article  CAS  PubMed  Google Scholar 

  28. Heljasvaara R, Nyberg P, Luostarinen J, Parikka M, Heikkila P, Rehn M, Sorsa T, Salo T, Pihlajaniemi T: Generation of biologically active endostatin fragments from human collagen XVIII by distinct matrix metalloproteases. Exp Cell Res 2005, 307: 292-304. 10.1016/j.yexcr.2005.03.021

    Article  CAS  PubMed  Google Scholar 

  29. Saarela J, Ylikarppa R, Rehn M, Purmonen S, Pihlajaniemi T: Complete primary structure of two variant forms of human type XVIII collagen and tissue-specific differences in the expression of the corresponding transcripts. Matrix Biol 1998, 16: 319-328. 10.1016/S0945-053X(98)90003-8

    Article  CAS  PubMed  Google Scholar 

  30. Valtola R, Salven P, Heikkila P, Taipale J, Joensuu H, Rehn M, Pihlajaniemi T, Weich H, deWaal R, Alitalo K: VEGFR-3 and its ligand VEGF-C are associated with angiogenesis in breast cancer. Am J Pathol 1999, 154: 1381-1390.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  31. Musso O, Theret N, Heljasvaara R, Rehn M, Turlin B, Campion JP, Pihlajaniemi T, Clement B: Tumor hepatocytes and basement membrane-producing cells specifically express two different forms of the endostatin precursor, collagen XVIII, in human liver cancers. Hepatology 2001, 33: 868-876. 10.1053/jhep.2001.23189

    Article  CAS  PubMed  Google Scholar 

  32. Elamaa H, Snellman A, Rehn M, Autio-Harmainen H, Pihlajaniemi T: Characterization of the human type XVIII collagen gene and proteolytic processing and tissue location of the variant containing a frizzled motif. Matrix Biol 2003, 22: 427-442. 10.1016/S0945-053X(03)00073-8

    Article  CAS  PubMed  Google Scholar 

  33. Wen W, Moses MA, Wiederschain D, Arbiser JL, Folkman J: The generation of endostatin is mediated by elastase. Cancer Res 1999, 59: 6052-6056.

    CAS  PubMed  Google Scholar 

  34. Lin HC, Chang JH, Jain S, Gabison EE, Kure T, Kato T, Fukai N, Azar DT: Matrilysin cleavage of corneal collagen type XVIII NC1 domain and generation of a 28-kDa fragment. Invest Ophthalmol Vis Sci 2001, 42: 2517-2524.

    CAS  PubMed  Google Scholar 

  35. Ferreras M, Felbor U, Lenhard T, Olsen BR, Delaisse J: Generation and degradation of human endostatin proteins by various proteinases. FEBS Lett 2000, 486: 247-251. 10.1016/S0014-5793(00)02249-3

    Article  CAS  PubMed  Google Scholar 

  36. Keane MP, Donnelly SC, Belperio JA, Goodman RB, Dy M, Burdick MD, Fishbein MC, Strieter RM: Imbalance in the expression of CXC chemokines correlates with bronchoalveolar lavage fluid angiogenic activity and procollagen levels in acute respiratory distress syndrome. J Immunol 2002, 169: 6515-6521.

    Article  CAS  PubMed  Google Scholar 

  37. Hamacher J, Lucas R, Lijnen HR, Buschke S, Dunant Y, Wendel A, Grau GE, Suter PM, Ricou B: Tumor necrosis factor-alpha and angiostatin are mediators of endothelial cytotoxicity in bronchoalveolar lavages of patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 2002, 166: 651-656. 10.1164/rccm.2109004

    Article  PubMed  Google Scholar 

  38. Richter AG, McKeown S, Rathinam S, Harper L, Rajesh P, McAuley DF, Heljasvaara R, Thickett DR: Soluble endostatin is a novel inhibitor of epithelial repair in idiopathic pulmonary fibrosis. Thorax 2009, 64: 156-161. 10.1136/thx.2008.102814

    Article  CAS  PubMed  Google Scholar 

  39. Pandya NM, Dhalla NS, Santani DD: Angiogenesis – a new target for future therapy. Vascul Pharmacol 2006, 44: 265-274. 10.1016/j.vph.2006.01.005

    Article  CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We would like to thank the staff and patients of the ICUs and the Wellcome Trust Clinical Research Facility. Mrs. Ritva Savilaakso and Jaana Peters are acknowledged for their excellent technical assistance. GDP was funded by the UK Intensive Care Society and is currently a NIHR clinician scientist. NN was funded by University Hospital Birmingham charities. RH was funded from the Finnish Centre of Excellence Programme (2000 to 2005) of the Academy of Finland. DRT is funded by the Wellcome Trust.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to David R Thickett.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

GDP, NN, SS, DM, AR, WT, MM and DRT recruited patients and performed bronchoscopy. GDP, NN and AR performed ELISA measurements. RH performed the western blot analyses. All authors contributed to writing the paper. GDP NN and DRT performed the statistical analyses. NN and GDP contributed equally to the paper.

Gavin D Perkins, Nazim Nathani and David R Thickett contributed equally to this work.

Authors’ original submitted files for images

Rights and permissions

Reprints and permissions

About this article

Cite this article

Perkins, G.D., Nathani, N., Richter, A.G. et al. Type XVIII collagen degradation products in acute lung injury. Crit Care 13, R52 (2009). https://doi.org/10.1186/cc7779

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/cc7779

Keywords