Human Surfactant Protein A Suppresses T Cell ... - ATS Journals

4 downloads 0 Views 351KB Size Report
Inflammation and Attenuates the Manifestations of Idiopathic. Pneumonia Syndrome in Mice. Shuxia Yang, Carlos Milla, Angela Panoskaltsis-Mortari, David H.
Human Surfactant Protein A Suppresses T Cell–Dependent Inflammation and Attenuates the Manifestations of Idiopathic Pneumonia Syndrome in Mice Shuxia Yang, Carlos Milla, Angela Panoskaltsis-Mortari, David H. Ingbar, Bruce R. Blazar, and Imad Y. Haddad Department of Pediatrics, Division of Pulmonary and Critical Care; Division of Bone Marrow Transplantation; Cancer Center; and Department of Medicine, University of Minnesota, Minneapolis, Minnesota

We have previously shown an association between growth factor–induced upregulation of surfactant protein (SP)-A and suppression of alveolar inflammation in our murine model of donor T cell–dependent lung dysfunction after bone-marrow transplantation, referred to as idiopathic pneumonia syndrome (IPS). We hypothesized that SP-A protects the lung in vivo from IPS injury by downregulation of alveolar inflammation. Human SP-A (100 ␮g), purified by n-butanol extraction or preparative isoelectric focusing, was transtracheally instilled on Day 4 after BMT during a time of in vivo donor T-cell activation. At 48 h after treatment, immunohistochemical staining of lung sections showed that SP-A did not alter T cell– dependent cellular infiltration. However, macrophages from SP-A–instilled mice were less injured and spontaneously produced less tumor necrosis factor-␣ than did cells from bufferinstilled mice. Although exogenous SP-A did not significantly alter bronchoalveolar lavage fluid (BALF) high levels of total protein (TP), an inverse correlation between BALF SP-A and TP concentrations (r ⫽ ⫺0.65; P ⫽ 0.02) was observed in SPA–treated but not in buffer-instilled mice. The only difference between the effects of the two sources of SP-A was that butanol-extracted SP-A, but not isoelectric focusing–purified SPA, suppressed the interferon-␥/nitric oxide pathway. We conclude that SP-A downregulates T cell–dependent alveolar inflammation by multiple pathways leading to decreased IPS injury.

Allogeneic (related and unrelated donor) bone-marrow (BM) transplantation (BMT) is a frequently used treatment for malignant, hematologic, immunologic, and genetic diseases. The success of this therapeutic modality is often compromised by the development of noninfectious diffuse lung injury referred to as idiopathic pneumonia syndrome (IPS). IPS occurs in 12 to 20% of all allogeneic BMT recipients, with a mortality rate in excess of 50% (1). Despite the profound morbidity and mortality associated with IPS, little is known about either etiology or treatment. Recent evidence shows that IPS injury likely results from persistent inflammation due to immune activation

(Received in original form November 8, 2000 and in revised form January 24, 2001) Address correspondence to: Imad Y. Haddad, M.D., University of Minnesota, Dept. of Pediatrics, 420 Delaware St. S.E., Minneapolis, MN 55455. E-mail: [email protected] Abbreviations: bronchoalveolar lavage fluid, BALF; bone marrow, BM; marrow cells supplemented with spleen cells, BMS; BM transplanation, BMT; cyclophosphamide, Cy; enzyme-linked immunosorbent assay, ELISA; interferon, IFN; immunoglobulin, Ig; idiopathic pneumonia syndrome, IPS; keratinocyte growth factor, KGF; lactic dehydrogenase, LDH; lipopolysaccharide, LPS; nitric oxide, NO; phosphate-buffered saline, PBS; sodium dodecyl sulfate, SDS; standard error, SE; surfactant protein, SP; total body irradiation, TBI; tumor necrosis factor, TNF; total protein, TP. Am. J. Respir. Cell Mol. Biol. Vol. 24, pp. 527–536, 2001 Internet address: www.atsjournals.org

and is potentiated by conditioning regimens (2–4). In our new murine IPS model that simulates the human condition, lung dysfunction in lethally irradiated BMT mice is dependent on infusion of allogeneic donor T cells (3). The most severe lung injury occurs in mice given allogeneic T cells and a commonly used conditioning regimen that includes cyclophosphamide (Cy). Lung dysfunction is associated with the activation of host alveolar macrophages and the production of large amounts of inflammatory mediators, including tumor necrosis factor (TNF)-␣ and nitric oxide (NO) (5). Surfactant protein (SP)-A, synthesized by alveolar type II and Clara cells, is strongly associated with pulmonary surfactant lipids (6). Therefore, it has been assumed that SP-A function is limited to surfactant homeostasis and surfactant resistance to inactivation by serum proteins. SP-A belongs to the collectin subgroup of the C-type lectin superfamily (7), and recent evidence indicates that its main function is in the first-line innate host defense against invading microbial agents. SP-A binds to specific carbohydrate structures present on the surface of bacteria and viruses. This binding initiates microbial aggregation and facilitates phagocytosis and killing by macrophages, monocytes, and other inflammatory cells (8, 9). In addition, SP-A has been shown to exhibit direct immunoregulatory functions on several types of immune cells. Both anti- and proinflammatory functions of SP-A have been reported (10–12). Although the method of SP-A isolation has been suggested as a potential explanation for the conflicting effects of SP-A, experimental evidence presented by Borron and colleagues shows that human SP-A purified either by n-butanol or preparative isoelectric focusing equally inhibits the proliferation of phytohemagutinin-activated human peripheral blood mononuclear cells (13). Harrod and associates (14) and Levine and coworkers (15) observed that the transtracheal injection of exogenous human SP-A prevents virus-induced inflammation, but whether the anti-inflammatory effects of SP-A are related to viral clearance versus direct downregulation of inflammation remains unclear. In the absence of infection, Borron and colleagues (11) reported that treatment of lipopolysaccharide (LPS)-exposed SP-A knockout mice with exogenous human SP-A inhibits LPS-induced production of TNF-␣ and NO. Because some LPS, albeit a small amount (8.6% of instilled LPS), was noted to bind SP-A, these authors could not exclude the possibility that this amount of LPS binding was sufficient to inhibit LPSinduced activation of inflammatory cells. Low levels of SP-A have been reported in patients with inflammatory lung diseases, including lung transplantation

528

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 24 2001

(16) and cystic fibrosis (17). Bronchoalveolar lavage fluid (BALF) SP-A and SP-D concentrations are significantly lower in patients at risk for adult respriatory distress syndrome (ARDS) and can predict those patients at high risk for severe ARDS and/or death after the onset of ARDS (18). Using our IPS model, we reported a correlation between air-space SP-A level and severity of alveolar inflammation (19). The effects of exogenous SP-A on donor T cell–dependent inflammation and subsequent development of lung dysfunction have not been thoroughly investigated. We hypothesized that SP-A treatment protects the lung in vivo from IPS injury by inhibiting allogeneic donor T cell–immune responses. Our results indicate that human SP-A purified from patients with alveolar proteinosis either by butanol extraction or preparative isoelectric focusing suppresses alveolar inflammation and attenuates the severity of IPS injury.

Materials and Methods BMT BMT was performed as previously described (3). Female B10.BR mice (H2k) (age 8 to 12 wk; Jackson Laboratory, Bar Harbor, ME) were given Cy (Cytoxan; Bristol Myers Squibb, Seattle, WA) 120 mg/kg/d as a conditioning regimen on Days ⫺3 and ⫺2 before BMT and were lethally total-body irradiated (7.5 Gray total body irradiation [TBI] by X-ray at a dose rate of 0.41 Gy/min) on the day before BMT. Donor female C57BL/6 (H2 b) (age 8 to 10 wk; Jackson Laboratory) BM was T cell–depleted with a monoclonal anti-Thy 1.2 antibody (clone 30-H-12, rat immunoglobulin [Ig]G2b, kindly provided by Dr. David Sachs, Massachusetts General Hospital, Boston, MA) plus complement (Neiffenegger Co., Woodland, CA). Recipient mice were transplanted via caudal vein with 20 ⫻ 106 C57BL/6 marrow cells supplemented with or without 15 ⫻ 106 spleen cells (BMS) as a source of IPS-causing T cells. Mice were housed in microisolator cages in the specific pathogen– free facility of the University of Minnesota and cared for according to the Research Animal Resources guidelines of that institution.

Injection of SP-A Human SP-A was obtained from the BALF of patients with alveolar proteinosis using two different methods for SP-A isolation. First, SP-A (labeled SP-Abutanol; provided by Dr. Jo Rae Wright, Duke University, Durham, NC) was purified by sequential extraction with n-butanol as previously described (20). The second source of SP-A (labeled SP-Aisoelectric; provided by Dr. David Phelps, Pennsylvania State University, Hershey, PA) was purified by preparative isoelectric focusing as previously described (21). SP-A was resuspended in 5 mM tris(hydroxymethyl)aminomethane (Tris), pH 7.4. Both SP-A preparations contained very low level of endotoxin (0.056 and 0.055 pg/␮g for SP-Abutanol and SP-Aisoelectric, respectively). Data generated from the two sources of SP-A were pooled except when differences between the effects of SP-A butanol and SP-Aisoelectric were encountered. On Day 4 after BMT (a time when donor T cells are being activated by host cells), mice were anesthetized with intraperitoneal sodium pentobarbital (5 mg) and injected transtracheally either with SP-A (100 ␮g) dispersed in 75 ␮l of 5 mM Tris or with an equal volume of sterile buffer, using a 27-gauge needle attached to a tuberculin syringe. TBI/Cy B10.BR mice given donor BM (BM ⫹Cy) without donor spleen T cells served as additional controls.

Bronchoalveolar Lavage Mice were killed on Day 6 after BMT, the time of peak allogeneic T-cell activation in vivo (22). The thoracic cavity was par-

tially dissected, and the trachea was cannulated with a 22-gauge angiocatheter and infused with 1 ml of ice-cold sterile phosphatebuffered saline (PBS) and withdrawn. This was repeated twice, and the return fluid was combined. A total of 10 ␮l of BALF was used to count the number of inflammatory cells and the remaining fluid was immediately centrifuged at 500 ⫻ g for 10 min at 4⬚C to pellet cells. BALF total protein was determined with the bicinchoninic acid (BCA) method, with bovine serum albumin (BSA) used as a standard. Lactic dehydrogenase (LDH) levels were measured by the colorimetric CytoTox 96 assay (Promega, Madison, WI), and LDH concentration (mU/liter) in the BALF was calculated using bovine heart LDH as standard.

Enzyme-Linked Immunosorbent Assay for SP-A SP-A concentration (␮g/ml) in cell-free BALF was determined by enzyme-linked immunosorbent assay (ELISA) using polyclonal rabbit antihuman SP-A (provided by Dr. David Phelps). The antihuman SP-A antibody also recognizes mouse SP-A (19). Equal aliquots (1 ␮l) of BALF were serially diluted using 50 mM Na2CO3-NaHCO3 buffer at pH 9.6, coated to ELISA plates, and allowed to bind for at least 18 h at 4 ⬚C. Nonspecific binding sites were blocked with 1% BSA for 1 h at room temperature. The wells were then incubated with the primary antibody (1:10,000 dilution) at 37⬚C for 1 h. Unbound antibody was removed by a series of washes with PBS-Tween 20 buffer. Horseradish peroxidase–conjugated goat antirabbit IgG (1:2,500 dilution; Sigma Co., St. Louis, MO) was added as the secondary antibody. After serial washes, color was developed by adding o-phenylenediaminedihydrochloride (Sigma Co.) and hydrogen peroxide to each well, and absorbance was read at 490 nm. Purified SP-A (0.125 to 3 ng), isolated from patients with alveolar proteinosis as previously described, was used as standard. Concentration of SP-A was calculated from each sample’s slope (absorbance/ ␮l BALF) and slope of standard curve (absorbance/ng SP-A). The lowest concentration of SP-A detectable by this method was approximately 0.2 ng/ ␮g of total protein (TP). To determine whether the presence of contaminating proteins in the BALF samples interfered with SP-A measurement we added a known amount of purified human SP-A (final concentration of 2 ␮g/ml) to solutions containing different concentrations of albumin (0, 0.5, 1, 2, 2.5, 5, 10, 50, and 100 mg/ml) and compared the accuracy of SP-A measurement by our ELISA method.

Western Blots for SP-A Equal volumes (20 ␮l) of BALF were solubilized in 0.1 M Tris buffer containing 50 ␮M dithiothreitol, 0.01% bromophenol blue, 1% sodium dodecyl sulfate (SDS), and 10% glycerol, and boiled for 5 min. The proteins were resolved by 12% SDS polyacrylamide gels, transferred to nitrocellulose paper, and probed with the antihuman SP-A antibody (1:10,000 dilution) followed by alkaline phosphatase–conjugated goat antirabbit IgG (1:7,500 dilution) as the secondary antibody. Bound antibody was detected with nitroblue tetrazolium and a 5-bromo-4-chloro-3indolyl-1phosphate kit (Sigma Co.). Additional experiments also showed that the SP-A antibody we used for SP-A quantification had similar affinities for both normal and nitrated/oxidized SP-A (data not shown).

Histology and Immunohistochemistry In one or two mice per group per experiment, a mixture of 1 ml optimal cutting temperature medium (OCT; Miles Laboratories, Inc., Elkhart, IN)/PBS (3:1) was infused in the trachea. The lung was snap-frozen in liquid nitrogen and stored at ⫺80⬚C. Frozen sections were cut 4 ␮m thick, mounted onto glass slides, and fixed for 5 min in acetone. Representative sections were stained with

Yang, Milla, Panoskaltsis-Mortari, et al.: Anti-inflammatory Effects of SP-A after Transplantation

hematoxylin and eosin (H&E) for histopathologic assessment. After a blocking step in 10% normal horse serum (Sigma Co.), sections were incubated for 30 min at 23 ⬚C with the following biotinylated monoclonal antibodies (PharMingen, San Diego, CA): anti-CD4 (clone GK1.5), anti-CD8 (clone 2.43), and anti–Mac-1 (cloneM1/70). Immunoperoxidase staining was performed using avidin-biotin blocking reagents, avidin-biotin (ABC)–peroxidase conjugate, and diaminobenzidine tetrahydrochloride (DAB) chromogenic substrate (Vector Laboratories, Inc., Burlingame, CA). The numbers of positive CD4 and CD8 cells in the lung were quantitated as the percent of nucleated cells at a magnification of 200 (⫻20 objective lens). Four fields per lung were evaluated.

Macrophage Cell Culture The BALF cell pellets from each treatment group were combined, washed twice in cold PBS, and resuspended in RPMI 1640 medium (Celox Laboratories, St. Paul, MN) containing 5% fetal calf serum (FCS), penicillin (100 U/ml), and streptomycin (100 ␮g/ml). Total cell number was determined with a hemacytometer, and cell viability was assessed by trypan blue exclusion. The quantity of 2 ⫻ 105 total viable cells/well was added to mouse IgG–coated, flat-bottomed, 96-well microtiter plates (Costar, Cambridge, MA), and macrophages/monocytes were allowed to adhere for 1 h at 37⬚C in 5% CO2 in air, followed by removal of unbound cells. The cells were maintained in culture at 37 ⬚C for 48 h in 5% CO2 in air. At termination of cell culture, supernatants were aspirated from individual culture wells for measurement of LDH, nitrite, and TNF-␣. Cells were washed twice with PBS and lysed with lysis solution (10⫻, Triton X-100; Promega), and cellular LDH release was measured. The percent cytotoxicity during culture of macrophages obtained from transplanted mice was calculated by dividing cell-free supernatant LDH by total cellular plus supernatant LDH of each well and multiplying by 100%. This method of measuring cytotoxicity has been demonstrated to be identical (within experimental error) to values determined in parallel 51chromium release assays (23). Total (supernatant plus cellular) LDH values were also used to evaluate for possible differences in adherent cell number between groups. Cell-free culture supernatant nitrite was measured by the Greiss reaction, and murine TNF-␣ was measured by ELISA with commercial kits (R&D Systems, Minneapolis, MN).

529

self-associate in the presence of calcium was determined as previously described (24). Sample and reference cuvettes contained SP-A (10 ␮g) dispersed in 100 mM saline and 5 mM Tris, pH 7.4. Aggregation was initiated by the addition of 5 mM calcium chloride. Aggregation was assessed by turbidity measurement at 400 nm using a Beckman spectrophotometer (Beckman Instruments, Fullerton, CA).

Statistical Analysis Results are expressed as means ⫾ standard error (SE). Data were analyzed by analysis of variance or Student’s t test. To evaluate for the presence of a correlation between BALF SP-A level and BALF TP or LDH in SP-A–instilled and buffer-instilled transplanted mice, Spearman’s rank correlation was used. Where a correlation was noted, multiple regression was used to determine the effect of SP-A treatment on the relationship between BALF SP-A and BALF TP/LDH levels. For this analysis, treatment group (treatment with SP-A or buffer) was entered as a covariate with SP-A level, as well as an interaction term representing the difference between the slopes for each group. For all analyses performed, a type I error probability of 0.05 was used as the cutoff for statistical significance.

Results Human SP-A Is Detected in the BALF of BMT Mice 48 h after Transtracheal Injection Using test samples containing a known amount of purified SP-A and increasing concentrations of proteins we first determined that, in the range of TP encountered in our BALF samples (0.25 to 1.25 mg/ml), the ELISA method using antihuman SP-A antibody accurately reflects SP-A content. At higher concentrations (⬎ 2.5 mg/ml), albumin progressively interfered with SP-A content (data not shown). ELISA (1 ␮l) of BALF from Day 6 after BMT demonstrated approximately a 2-fold increase in SP-A level in

T-Cell Culture Approximately 40% of BALF cells from donor T cell–recipient mice are alloactivated T cells (5). Nonadherent BALF cells ( ⬎ 90% T cells) of SP-A–instilled or buffer-instilled BMT mice were pooled. Cells were counted using a hemacytometer and cell viability was assessed by trypan blue exclusion. The quantity of 5 ⫻ 105 cells was suspended in minimal essential medium (Celox) containing 10% FCS, 10 mM N-2-hydroxyethylpiperazine-N⬘ethane sulfonic acid buffer, 1 mM sodium pyruvate, and amino acid supplements (1.5 mM L-glutamine, L-arginine, and L-asparagine; Sigma) in flat-bottomed 96-well sterile plates. After 24 h in culture, spontaneous murine interferon (IFN)- ␥ production was measured in the cell-free culture supernatant by ELISA (R&D Systems).

Physical State of Purified Human SP-A To determine differences in the oligomerization state of the two SP-A preparations used for treatment of mice, Western blots of SP-Abutanol and SP-Aisoelectric were performed under reducing and nonreducing conditions. Purified SP-A (1 ␮g) was solubilized and boiled for 5 min. The proteins were resolved by 12% SDS polyacrylamide gels, transferred to nitrocellulose paper, and probed with the antihuman SP-A antibody (1:10,000 dilution) as described earlier. In addition the ability of SP-A butanol and SP-Aisoelectric to

Figure 1. SP-A treatment increases BALF SP-A levels. ELISA for SP-A (␮g/ml) using equal volume (1 ␮l) of BALF 48 h after transtracheal injection of human SP-A (100 ␮g) or equal volume of sterile buffer. Cy/TBI-conditioned B10.BR recipient mice were given C57BL/6 BM, (BM⫹Cy) with donor spleen T cells from C57BL/6 mice (BMS⫹Cy). Butanol-extracted or isoelectric focusing purified human alveolar proteinosis SP-A or buffer were injected on Day 4 after BMT. The primary antibody used for ELISA was the polyclonal rabbit antihuman SP-A antibody followed by horseradish peroxidase–conjugated goat antirabbit IgG as the secondary antibody. Return BALF volume was the same in all groups. Values are means ⫾ SE obtained from eight to 12 mice in each group. *P ⬍ 0.05 compared to control (BM⫹Cy).

530

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 24 2001

Figure 2. Western blot of equal BALF volume (20 ␮l) obtained 48 h after transtracheal injection of human SP-A (100 ␮g; lanes 2–4), or equal volume of sterile buffer (lanes 5–8) on Day 4 after BMT into Cy/TBI conditioned B10.BR mice given BM and donor spleen T cells from C57BL/6 mice (BMS⫹Cy). SP-A was detected with a polyclonal rabbit antisheep SP-A antibody that cross-reacts with mouse SP-A. Lane 1 represents purified human SP-A (1 ␮g). Shown is a representative blot using butanol-extracted SP-A, which was repeated once. Blot using isoelectric focusing SP-A showed similar results.

the BALF of mice 48 h after exogenous human SP-A injection as compared with nontreated controls (Figure 1). The presence of the SP-A dimer (ⵑ 66 kD) on Western blot of BALF (20 ␮l) from SP-A–treated compared with buffer-instilled TBI/Cy-conditioned mice given allogeneic T cells (BMS⫹Cy) indicates that some of the recovered SP-A was of human origin (SP-A from patients with alveolar proteinosis cannot be completely reduced) (Figure 2). The additional band at ⵑ 43 kD observed in all samples may represent SP-D. Our data indicate that exogenous human SP-A was incompletely cleared from mouse lung after 48 h of administration. The results demonstrate that BALF inflammatory and immune cells were exposed to increased levels of SP-A during the critical in vivo cellactivation period. Inverse Correlation between BALF SP-A Level and Indices of Lung Injury in SP-A–Treated BMS⫹Cy Mice The percent recovery of BALF was similar in all groups of mice (⬎ 90% of instilled volume). Exogenous SP-A did not significantly alter the high levels of TP and LDH (indices of permeability edema and cytolysis) in Cy/TBI donor T cell–recipient mice (BMS⫹Cy) analyzed on Day 6 after BMT (Table 1). However, we observed that the BALF of SP-A–treated mice with the highest concentration of SP-A contained the lowest level of TP and LDH. A significant negative correlation between BALF SP-A and TP existed in the group of mice treated with human SP-A (r ⫽ ⫺0.65, P ⫽ 0.02) (Figure 3). In contrast, for buffer-instilled mice a

positive correlation between BALF SP-A and TP was observed (r ⫽ 0.53, P ⫽ 0.05). The positive correlation noted between BALF SP-A and TP in buffer-instilled mice may be attributed to occasional low return yield of all components of the alveolar space in some samples. The same trends were noted in the BALF SP-A and LDH levels—a negative correlation in SP-A–treated mice and a positive correlation in buffer-instilled mice—however these correlations did not reach statistical significance (r ⫽ ⫺0.37 and 0.2, respectively; P ⬎ 0.2) (Figure 3). By multiple regression analysis, the change in the correlation between BALF SP-A and TP/LDH according to treatment group (SP-A or buffer treatment) was highly significant for TP (P ⫽ 0.006) and marginally significant for LDH (P ⫽ 0.10). These data provide evidence that the ability to achieve high alveolar SP-A levels was associated with attenuated indices of lung injury. SP-A Treatment Does Not modify BMS⫹Cy Mice BALF and Lung Cellularity Injection of allogeneic T cells in Cy/TBI BMT mice increased the number of inflammatory cells recovered in return Day 6 after BMT BALF, and SP-A treatment on Day 4 did not significantly modify the number of BALF cells (Table 1). Lung tissues were harvested on Day 6 after BMT. Consistent with our published data (3), histologic assessment of lungs from Cy/TBI T cell–recipient mice (BMS⫹Cy) exhibited infiltration with inflammatory cells (Figure 4) that were mainly T cells and macrophages/monocytes, as assessed by immunohistochemistry (Figure 5). Treatment with SP-A on Day 4 after BMT did not significantly alter the number or composition of lung-infiltrating cells; before and after SP-A treatment, respectively, amounts were: % CD4⫹ cells, 11.6 ⫾ 2.4 and 10.2 ⫾ 2.3; % CD8⫹ cells, 12.3 ⫾ 4.4 and 9.1 ⫾ 3.8; % Mac-1⫹ cells, 35 ⫾ 4 and 28 ⫾ 5. Data are expressed as percent of nucleated cells expressing the surface marker in the lung as determined by counting four fields per lung section under light microscopy. Values are means ⫾ SE obtained from two mice per group per experiment. Two experiments were performed: butanol-extracted SP-A was used in the first, and SP-A purified by preparative isoelectric focusing in the second. SP-A Treatment Prevents Cytotoxicity of Macrophages from BMS⫹Cy Mice We have shown that macrophages obtained from inflamed and injured lungs undergo autolysis when cultured in vitro, and this lysis in culture reflects the toxic environment from

TABLE 1

Effects of SP-A on BALF indices of lung injury and cellularity BALF

Total protein (mg/ml) LDH (mU/liter) Number of cells ⫻ 104/ml

BM⫹Cy

BMS⫹Cy⫹I.T. buffer

BMS⫹Cy⫹I.T. SPA

0.25 ⫾ 0.03 84 ⫾ 8 3.5 ⫾ 1.2

0.51 ⫾ 0.02* 225 ⫾ 9* 14.5 ⫾ 3.2*

0.54 ⫾ 0.06* 199 ⫾ 16* 11.8 ⫾ 4.1*

Values are means ⫾ SE for n ⫽ 8–12 mice/group. BM⫹Cy: BMT⫹TBI⫹Cy (120 mg/kg/d on Days ⫺3 and ⫺2 before BMT); BMS⫹Cy: BMT⫹TBI⫹donor spleen T cells⫹Cy. Human SP-A (100 ␮g) or buffer was instilled intratracheally (I.T.) on Day 4 after BMT. Bronlchoalveolar lavage was performed 48 h after SP-A treatment. SP-A was isolated from the BALF of alveolar proteinosis patients using butanol extraction and preparative isoelectric focusing. *P ⬍ 0.05 compared with control (BM⫹Cy).

Yang, Milla, Panoskaltsis-Mortari, et al.: Anti-inflammatory Effects of SP-A after Transplantation

531

of supernatant and cellular LDH levels after 48 h in culture as described in MATERIALS AND METHODS. In vivo SPA treatment partially prevented the percent cytotoxicity of macrophages/monocytes obtained from TBI/Cy-conditioned mice given allogeneic T cells (BMS⫹Cy) (Figure 6). SP-A Treatment Downregulates the Activation of Macrophages from BMS⫹Cy Mice Total (supernatant plus cellular) LDH level/well confirmed approximately equal numbers of cells in the wells (data not shown). Cultured macrophages/monocytes from Day 6 after BMT BALF of TBI/Cy mice given allogeneic T cells (BMS⫹Cy) spontaneously produced large amounts of TNF-␣ and NO (Figure 7). TNF-␣ and NO production was dependent on the infusion of allogeneic T cells because macrophages from TBI/Cy mice not given donor T cells (BM⫹Cy) did not spontaneously generate detectable TNF-␣ and NO during the 48-h culture period (Figure 7). Treatment with SP-A isolated by either n-butanol extraction (SP-Abutanol) or preparative isoelectric focusing (SP-Aisoelectric) suppressed the spontaneous high level of TNF-␣ produced by macrophages obtained from BMS⫹Cy mice (Figures 7A and 7C). However, although macrophages from Cy/TBI donor T cell–recipient mice treated with SP-Abutanol also generated less nitrite (Figure 7B), cells from BMS⫹Cy mice instilled with SP-Aisoelectric persisted to spontaneously generate large amounts of NO (Figure 7D).

Figure 3. Correlation between SP-A and TP/LDH levels in the BALF obtained 48 h after SP-A/buffer instillation. SP-A was extracted from therapeutic BALF of patients with alveolar proteinosis using butanol extraction or isoelectric focusing. Transtracheal SP-A (100 ␮g) or equal volume of sterile buffer (75 ␮l) was injected on Day 4 after BMT into Cy/TBI-conditioned B10.BR mice given C57BL/6 BM plus donor spleen T cells from C57BL/6 mice (BMS⫹Cy). (A) In SP-A–treated mice (filled circles), a negative correlation was observed between BALF SP-A and TP levels (solid line, r ⫽ ⫺0.65, P ⫽ 0.02). In contrast, a positive correlation was observed in buffer-instilled mice (open circles) between BALF SP-A and TP levels (broken line, r ⫽ 0.53, P ⫽ 0.05). (B) In SP-A–treated mice ( filled triangles), a negative correlation and a positive correlation was seen in buffer-instilled mice (open triangles) between BALF SP-A and LDH levels (r ⫽ ⫺0.37 and 0.2, respectively; P ⬎ 0.2 for both). By multiple regression analysis, the change in the direction of the correlation by SP-A treatment was significant for TP (P ⫽ 0.006) and marginally significant for LDH (P ⫽ 0.10). BALF SP-A was measured by ELISA using antihuman SP-A antibody. TP was measured by the BCA method and LDH by colorimetric assay (CytoTox 96).

which the cells were extracted (25). BALF cells obtained from BMT mice 48 h after transtracheal buffer or SP-A injection (Day 6 after BMT) were allowed to adhere on flatbottomed 96-well microtiter plates. Examination of adherent cells by Wright-Giemsa staining demonstrated that ⬎ 95% of cells were macrophages (data not shown), and showed approximately equal numbers of cells/well. Percent cytotoxicity of cells was calculated by measurement

Treatment with SP-Abutanol, but Not SP-Aisoelectric, Suppresses IFN-␥ Production by Lung-Infiltrating T Cells from BMS⫹Cy Mice Because activated T cell–derived IFN-␥ is the most potent stimulus for inducible NO synthase expression and highoutput NO generation, we evaluated IFN-␥ production by cultured BALF T cells. At 24 h, T cells obtained from BALF of Cy/TBI mice injected with allogeneic T cells and instilled with SP-Abutanol, but not SP-Aisoelectric, produced less IFN-␥ compared with cells from buffer-instilled BMS⫹Cy mice (Figure 8). Taken together, the data suggest that SP-A downregulates donor T cell–dependent activation of alveolar macrophages by multiple mechanisms. Effects of Isolation Method on Physical Properties of SP-A Western blot under reducing and nonreducing conditions did not reveal differences in the state of oligomerization between SP-Abutanol and SP-Aisoelectric (Figure 9). Similarly, there was no difference between the two preparations of SP-A in the ability to self-aggregate in the presence of calcium (data not shown). These data indicate that purification of alveolar proteinosis SP-A by butanol and preparative isoelectric focusing yields a protein of similar state of oligomerization and self-aggregation, and that this physical state is not the reason for the differences in the antiinflammatory effects of SP-Abutanol and SP-Aisoelectric.

Discussion The major findings in this study are that human alveolar proteinosis SP-A instilled into the lung of Cy/TBI–conditioned allogeneic T cell–recipient mice on Day 4 after BMT suppresses donor T cell–dependent inflammation and par-

532

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 24 2001

Figure 4. Histologic assessment of inflammation on Day 6 after BMT, 48 h after transtracheal injection of SP-A/buffer. H&E-stained frozen sections of lungs taken from Cy/TBI B10.BR mice given BM from C57BL/6 (BM⫹Cy; A), Cy/TBI BMT mice given C57BL/6 spleen T cells and instilled with 75 ␮l of 5 mM Tris buffer (BMS⫹Cy; B), and Cy/TBI BMT mice given donor spleen T cells and instilled with human SP-A (100 ␮g) (BMS⫹Cy; C). Lung infiltration with inflammatory cells was dependent on infusion of donor T cells. The histologic appearance of lung sections from SP-A–treated mice was similar to that of buffer-instilled mice. Many of the inflammatory cells were macrophages/monocytes as determined by staining lung sections with Mac-1 biotinylated monoclonal antibody (insert D). (Original magnification: ⫻100; resolution power: ⫻40 objective lens). Shown is a representative figure using butanol-extracted SP-A with similar results obtained when preparative isoelectric SP-A was instilled.

tially attenuates permeability edema and cytolysis associated with IPS injury. Because allogeneic T cells are critical for development of lung dysfunction in our model (3), and T cell–immune responses have been implicated in IPS pathogenesis (26, 27), the most likely mechanism for the protective effects of SP-A is downregulation of T cell– dependent inflammation and subsequent T cell–induced

stimulation of host alveolar macrophages and lung-infiltrating monocytes. The clearance half-life of murine SP-A from the air spaces has been calculated to be 10.2 h (28). Therefore, we expected to find low levels of exogenous SP-A in the BALF collected 48 h after exogenous SP-A injection. Instead, we noted a 2-fold higher SP-A concentration in the BALF of

Figure 5. Effect of transtracheal SP-A on lung-infiltrating T cells by immunoperoxidase staining of lung sections from the indicated group of mice. Frozen lung sections taken on Day 6 after BMT were incubated with biotinylated monoclonal anti-CD4 and anti-CD8 antibodies and developed with peroxidase-conjugated ABC and DAB chromogen (methyl green counterstain). Original magnification: ⫻100; resolution power equivalent to ⫻40 objective lens.

Yang, Milla, Panoskaltsis-Mortari, et al.: Anti-inflammatory Effects of SP-A after Transplantation

Figure 6. SP-A treatment prevents cytotoxicity of macrophages obtained from mice on Day 6 after BMT. BALF macrophages from Cy/TBI-conditioned B10.BR recipient mice given C57BL/6 bone marrow (BM⫹Cy) and spleen T cells (BMS⫹Cy) were cultured for 48 h as described in MATERIALS AND METHODS. BMS⫹Cy mice were instilled transtracheally with human SP-A or buffer on Day 4 after BMT. Percent cytotoxicity was calculated by dividing LDH level in cell-free supernatant by total LDH (supernatant plus cellular) in each well, then multiplying by 100. Equal numbers of cells per well was confirmed by light microscopy and total LDH measurement. Values are means ⫾ SE obtained from two experiments with five wells of pooled macrophages obtained from six mice per group per experiment. SP-A was extracted by butanol in one experiment and preparative isoelectric focusing in a second experiment. n.d., not determined. *P ⬍ 0.05 compared with BM⫹Cy. +P ⬍ 0.05 comparing the effect of SP-A treatment in BMS⫹Cy mice.

533

SP-A–treated mice compared with that in buffer-instilled mice. Potential explanations for persistence of human SP-A in air spaces of mice are recycling of the exogenous surfactant or inefficient uptake of human SP-A by murine alveolar type II cells during IPS injury. Although treatment of Cy/TBI–conditioned T cell–recipient mice with transtracheal SP-A did not significantly decrease BALF TP and LDH levels, we noted that exogenous SP-A had an effect on the relationship between BALF SP-A and TP/LDH levels. SP-A treatment reversed the direction of the correlation between BALF SP-A level and TP from positive to negative. Higher SP-A level in the air spaces was associated with lower TP and decreased permeability edema, providing strong evidence for an in vivo tissue-protective effect for SP-A. Although a similar trend between BALF SP-A and LDH levels was observed, the variability in LDH prevented significance of the inverse correlation between SP-A and LDH in SP-A–treated mice. The hypothesis that SP-A treatment prevents cytolysis is strengthened by the finding that macrophages from SP-A–treated mice given allogeneic T cells are significantly less injured compared with cells from buffer-instilled mice. These findings are consistent with the recent study of Spech and coworkers showing that human SP-A prevents silica-induced toxicity of cultured rat alveolar macrophages (29). Several groups have shown that SP-A inhibits the activation of a variety of inflammatory cell-types. Borron and associates demonstrated that SP-A obtained by different

Figure 7. SP-A treatment downregulates macrophage activation obtained from mice on Day 6 after BMT. BALF macrophages from Cy/TBI conditioned B10.BR recipient mice given C57BL/6 BM (BM⫹Cy) and spleen T cells (BMS⫹Cy) were cultured for 48 h as described in MATERIALS AND METHODS. BMS⫹Cy mice were instilled transtracheally with human SP-A or buffer on Day 4 after BMT. At termination of cell culture, supernatants were aspirated from individual culture wells for nitrite (Greiss reaction), and murine TNF-␣ (ELISA). Butanol-extracted SP-A (SPAbutanol) suppressed the spontaneous macrophage-derived TNF-␣ and NO production (A and B). SP-A extracted by preparative isoelectric focusing (SP-Aisoelectric) suppressed the spontaneous TNF-␣ (C), but not NO production (D). Equal numbers of cells per well was confirmed by light microscopy and total LDH measurement. Values are means ⫾ SE obtained from at least five wells of pooled macrophages obtained from six mice/group. *P ⬍ 0.05 compared with control (BM⫹Cy). +P ⬍ 0.05 comparing the effect of SP-A treatment in BMS⫹Cy group.

534

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 24 2001

Figure 8. Butanol-extracted SP-A, but not SP-A purified by preparative isoelectric focusing, suppresses spontaneous IFN-␥ production by T cells from Cy/TBI-conditioned B10.BR recipient mice given C57BL/6 BM and spleen T cells (BMS⫹Cy). BMS⫹Cy mice were instilled transtracheally with human SP-A or buffer on Day 4 after BMT. T cells were obtained from Day 6 after BMT nonadherent BALF cells as described in MATERIALS AND METHODS and cultured for 24 h at 37⬚C. Murine IFN-␥ was measured in cell-free culture medium using ELISA. Only butanol-extracted SP-A suppressed the spontaneous IFN-␥ production by alloactivated T cells. Values are means ⫾ SE obtained from three wells of pooled nonadherent BALF cells obtained from six mice/group. *P ⬍ 0.05 comparing the effect of instilled SP-A in each group.

isolation methods inhibits mitogen-activated human T-cell proliferation and interleukin-2 production (13). Wang and colleagues confirmed these findings using non–butanol extracted SP-A, concluding that SP-A suppresses lymphocyte proliferation and allergen-induced histamine release (30). Similarly, SP-A isolated by different methods inhibits TNF-␣ production by alveolar macrophages/monocytes stimulated with LPS (31) or Candida albicans (10). Our results indicate that treatment with human SP-A in vivo suppresses donor T cell–immune responses and T cell–dependent macrophage activation. SP-A may inhibit T-cell proliferation and activation by binding to specific cell-surface receptors. For example, the addition of a polyclonal antibody against an SP-A receptor (SP-R210) completely blocked the inhibition of T-cell proliferation by SP-A (13). SP-A may also interfere with the process of antigen presentation by the blocking of a costimulatory signal crucial for T-cell activation. These costimulatory molecules often contain mannose-linked oligosaccharides to which SP-A is known to have high affinity (30). In addition, SP-A has been shown to activate phagocytic cell signaling pathways, including induction of tyrosine phosphorylation (32), and modifications of intracellular calcium concentrations (33). The basic mechanisms for the anti-inflammatory effects of SP-A in our in vivo model have yet to be elucidated. Although human alveolar proteinosis SP-A extracted by two different methods downregulates alveolar inflammation and prevents the manifestation of IPS, our results suggest that the method of SP-A purification may have an effect on the main anti-inflammatory pathway. Treatment of BMS⫹Cy mice with SP-A purified by butanol extraction suppressed IFN-␥ production by alloactivated lung-infiltrating T cells and, therefore, T cell–induced macrophagederived production of NO and TNF-␣. A direct effect of

Figure 9. Western blot showing a comparison of the oligomerization state of human alveolar proteinosis SP-A isolated by butanol precipitation and preparative isoelectric focusing. The two preparations of SP-A were resolved by 12% SDS polyacrylamide gels under reducing and nonreducing conditions as indicated in the figure. The proteins were transferred to nitrocellulose paper and probed with antihuman SP-A antibody, followed by alkaline phosphatase-conjugated goat antirabbit IgG. Shown is a representative blot which was repeated twice with similar results.

SP-A on macrophages cannot be excluded. Treatment with SP-A purified by preparative isoelectric focusing also suppressed macrophage-derived TNF-␣, but not NO, and did not modify IFN-␥ production by T cells. In an attempt to understand the reasons for the differential inhibitory effects of SP-Abutanol and SP-Aisoelectric we compared the state of oligomerization and self-aggregation in the presence of calcium. Human alveolar proteinosis SP-A is known to be more aggregated than normal SP-A (34). Butanol extraction can inactivate dog and rat SP-A, but did not affect the ability of human alveolar proteinosis SP-A to stimulate the production of oxidants by alveolar macrophages (24). We were unable to detect differences between the physical properties of SP-Abutanol and SP-Aisoelectric as assessed by the ability to self-aggregate in the presence of calcium and by Western blotting. Because the two preparations of SP-A were compared between experiments, we cannot rule out that interexperimental variations in the state of activation of lung-infiltrating immune cells contributed to the observed differences in the anti-inflammatory effects of SP-Abutanol and SP-Aisoelectric. In contrast to our data, other groups have shown that SP-A stimulates cytokine and NO production by lymphocytes and macrophages/monocytes (35). Kremlev and coworkers reported that human alveolar proteinosis SP-A purified by preparative isoelectric focusing stimulates cytokine production by human peripheral blood mononuclear cells and by a monocytic cell line (12, 36). Although the reason(s) for the conflicting effects of SP-A on inflammation may include contamination of SP-A preparations with LPS and the cell-culture model used (reviewed in refs. 37 and 38), it is noteworthy that all studies performed in vivo (including our study) thus far have shown an antiinflammatory role of SP-A (11, 14). SP-A is known to protect the host from bacterial infection by promoting microbial uptake by alveolar macrophages and enhancing killing by stimulating the production of oxygen radicals (9). These data may seem contradictory to the anti-inflammatory role of SP-A. We propose that in the context of infection, SP-A protects the host by stimulating the innate defense system. However, if activation of the more potent and specific adaptive immune system occurs, SP-A acts to protect the delicate tissues of the lung from T cell–immune responses. SP-A instilled on Day 4 after BMT did not modify the number or composition of BALF or lung-infiltrating inflammatory cells. Although the signals for regulating lung

Yang, Milla, Panoskaltsis-Mortari, et al.: Anti-inflammatory Effects of SP-A after Transplantation

inflammatory cell recruitment may have been released before SP-A instillation, our results are in agreement with the findings of Borron and colleagues showing that human SP-A instilled in LPS-challenged mice inhibited alveolar inflammation without altering the number of inflammatory cells recovered in the BALF (11). These findings are also consistent with our recent report that the pre-BMT systemic administration of keratinocyte growth factor (KGF), a mediator of lung repair, upregulates endogenous SP-A expression and suppresses T cell–dependent inflammation without modifying total BALF cellularity or composition (19). McIntosh and colleagues suggested that SP-A is a lungspecific acute phase protein by showing upregulation of endogenous SP-A messenger RNA and protein expression after transtracheal injection of LPS (39). However, during intense inflammation the generation of severe oxidative/ nitrative stress damages SP-A–producing cells and prevents SP-A upregulation. Similarly, we have shown that the generation of oxidative/nitrative stress in lungs of Cy/ TBI mice given donor T cells is associated with impaired ability of pre-BMT KGF to upregulate SP-A and prevent lung inflammation (19). A bolus injection of SP-A, as shown herein, may limit the tissue-damaging production of inflammatory mediators. Our study strengthens the rationale for using SP-A–containing surfactant preparations for the treatment of IPS and, potentially, other inflammatory lung diseases. In summary, these studies demonstrate that enhancing alveolar levels of SP-A during donor T cell–dependent inflammation suppresses the activation of lung-infiltrating inflammatory cells, and attenuates permeability edema and cytolysis. The protective effects of SP-A may include inhibitory effects on the activation of a variety of cell types. The effects of exogenous SP-A on survival in this IPS model have yet to be elucidated. Acknowledgments: This work was supported by grants from the American Lung Association (JM-CIA), American Heart Association (Minnesota Affiliate, Inc.), the Viking Children’s Fund, and NIH R0-1 HL55209, HL67334, and AI34495. The authors thank Dr. Jo Rae Wright and Dr. David Phelps for providing SP-A and SP-A antibodies. The authors gratefully acknowledge the expert technical assistance of John Bob Hermanson.

References 1. Clark, J. G., J. A. Hansen, M. I. Hertz, R. Parkman, L. Jensen, and H. H. Peavy. 1993. Idiopathic pneumonia syndrome after bone marrow transplantation. Am. Rev. Respir. Dis. 147:1601–1606. 2. Clark, J. G., D. K. Madtes, T. R. Martin, R. C. Hackman, A. L. Farrand, and S. W. Crawford. 1999. Idiopathic pneumonia after bone marrow transplantation: cytokine activation and lipopolysaccharide amplification in the bronchoalveolar compartment. Crit. Care Med. 27:1800–1806. 3. Panoskaltsis-Mortari, A., P. A. Taylor, T. M. Yaeger, O. D. Wangensteen, P. B. Bitterman, D. H. Ingbar, D. A. Vallera, and B. R. Blazar. 1997. The critical early proinflammatory events associated with idiopathic pneumonia syndrome in irradiated murine allogeneic recipients are due to donor T cell infusion and potentiated by cyclophosphamide. J. Clin. Invest. 100: 1015–1027. 4. Clark, J. G., D. K. Madtes, R. C. Hackman, W. Chen, M. A. Cheever, and P. J. Martin. 1998. Lung injury induced by alloreactive Th1 cells is characterized by host-derived mononuclear cell inflammation and activation of alveolar macrophages. J. Immunol. 161:1913–1920. 5. Haddad, I. Y., A. Panoskaltsis-Mortari, D. H. Ingbar, S. Yang, C. E. Milla, and B. R. Blazar. 1999. High levels of peroxynitrite are generated in the lungs of irradiated mice given cyclophosphamide and allogeneic T cells: a potential mechanism of injury after marrow transplantation. Am. J. Respir. Cell Mol. Biol. 20:1125–1135.

535

6. Tino, M. J., and J. R. Wright. 1996. Surfactant protein A stimulates phagocytosis of specific pulmonary pathogens by alveolar macrophages. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 270:L677–L688. 7. Thiel, S. and K. B. Reid. 1989. Structures and functions associated with the group of mammalian lectins containing collagen-like sequences. FEBS Lett. 250:78–84. 8. Hartshorn, K. L., E. Crouch, M. R. White, M. L. Colamussi, A. Kakkanatt, B. Tauber, V. Shepherd, and K. N. Sastry. 1998. Pulmonary surfactant proteins A and D enhance neutrophil uptake of bacteria. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 274:L958–L969. 9. LeVine, A. M., K. E. Kurak, J. R. Wright, W. T. Watford, M. D. Bruno, G. F. Ross, J. A. Whitsett, and T. R. Korfhagen. 1999. Surfactant protein-A binds group B streptococcus enhancing phagocytosis and clearance from lungs of surfactant protein-A-deficient mice. Am. J. Respir. Cell Mol. Biol. 20:279–286. 10. Rosseau, S., P. Hammerl, U. Maus, A. Gunther, W. Seeger, F. Grimminger, and J. Lohmeyer. 1999. Surfactant protein A down-regulates proinflammatory cytokine production evoked by Candida albicans in human alveolar macrophages and monocytes. J. Immunol. 163:4495–4502. 11. Borron, P., J. C. McIntosh, T. R. Korfhagen, J. A. Whitsett, J. Taylor, and J. R. Wright. 2000. Surfactant-associated protein A inhibits LPS-induced cytokine and nitric oxide production in vivo. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 278:L840–L847. 12. Kremlev, S. G., T. M. Umstead, and D. S. Phelps. 1997. Surfactant protein A regulates cytokine production in the monocytic cell line THP-1. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 272:L996–L1004. 13. Borron, P., F. X. McCormack, B. M. Elhalwagi, Z. C. Chroneos, J. F. Lewis, S. Zhu, J. R. Wright, V. L. Shepherd, F. Possmayer, K. Inchley, et al. 1998. Surfactant protein A inhibits T cell proliferation via its collagen-like tail and a 210-kDa receptor. Am. J. Physiol. 275:L679–L686. 14. Harrod, K. S., B. C. Trapnell, K. Otake, T. R. Korfhagen, and J. A. Whitsett. 1999. SP-A enhances viral clearance and inhibits inflammation after pulmonary adenoviral infection. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 277:L580–L588. 15. LeVine, A. M., J. Gwozdz, J. Stark, M. Bruno, J. Whitsett, and T. Korfhagen. 1999. Surfactant protein-A enhances respiratory syncytial virus clearance in vivo. J. Clin. Invest. 103:1015–1021. 16. Casals, C., A. Varela, M. L. Ruano, F. Valino, J. Perez-Gil, N. Torre, E. Jorge, F. Tendillo, and J. L. Castillo-Olivares. 1998. Increase of C-reactive protein and decrease of surfactant protein A in surfactant after lung transplantation. Am. J. Respir. Crit. Care Med. 157:43–49. 17. Postle, A. D., A. Mander, K. B. Reid, J. Y. Wang, S. M. Wright, M. Moustaki, and J. O. Warner. 1999. Deficient hydrophilic lung surfactant proteins A and D with normal surfactant phospholipid molecular species in cystic fibrosis. Am. J. Respir. Cell Mol. Biol. 20:90–98. 18. Greene, K. E., J. R. Wright, K. P. Steinberg, J. T. Ruzinski, E. Caldwell, W. B. Wong, W. Hull, J. A. Whitsett, T. Akino, Y. Kuroki. et al. 1999. Serial changes in surfactant-associated proteins in lung and serum before and after onset of ARDS. Am. J. Respir. Crit. Care Med. 160:1843–1850. 19. Yang, S., A. Panoskaltsis-Mortari, D. H. Ingbar, S. Matalon, S. Zhu, E. R. Resnik, C. L. Farrell, D. L. Lacey, B. R. Blazar, and I. Y. Haddad. 2000. Cyclophosphamide prevents systemic keratinocyte growth factor-induced upregulation of surfactant protein A after allogeneic transplant in mice. Am. J. Respir. Crit. Care Med. 62:1884–1890. 20. Wright, J. R., J. D. Borchelt, and S. Hawgood. 1989. Lung surfactant apoprotein SP-A (26-36 kDa) binds with high affinity to isolated alveolar type II cells. Proc. Natl. Acad. Sci. USA 86:5410–5414. 21. Wang, G., D. S. Phelps, T. M. Umstead, and J. Floros. 2000. Human SP-A protein variants derived from one or both genes stimulate TNF-alpha production in the THP-1 cell line. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 278:L946–L954. 22. Blazar, B. R., P. A. Taylor, A. Panoskaltsis-Mortari, and D. A. Vallera. 1998. Rapamycin inhibits the generation of graft-versus-host disease and graft-versus-leukemia-causing T cells by interfering with the production of Th1 or Th1 cytotoxic cytokines. J. Immunol. 160:5355–5365. 23. Behl, C., J. B. Davis, R. Lesley, and D. Schubert. 1994. Hydrogen peroxide mediates amyloid B protein toxicity. Cell 77:817–827. 24. Van Iwaarden, J. F., F. Teding van Berkhout, J. A. Whitsett, R. S. Oosting, and L M. van Golde. 1995. A novel procedure for the rapid isolation of surfactant protein A with retention of its alveolar-macrophage-stimulating properties. Biochem. J. 309:551–555. 25. Haddad, I. Y., A. Panoskaltsis-Mortari, D. H. Ingbar, E. R. Resnik, S. Yang, C. L. Farrel, D. L. Lacey, D. N. Cornfield, and B. R. Blazar. 1999. Interactions of keratinocyte growth factor with a nitrating species after marrow transplantation in mice. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 277:L391–L400. 26. Cooke, K. R., W. Krenger, G. Hill, T. R. Martin, L. Kobzik, J. Brewer, R. Simmons, J. M. Crawford, M. R. van den Brink, and J. L. Ferrara. 1998. Host reactive donor T cells are associated with lung injury after experimental allogeneic bone marrow transplantation. Blood 92:2571–2580. 27. Crawford, S. W., G. Longton, and R. Storb. 1993. Acute graft-versus-host disease and the risks for idiopathic pneumonia after marrow transplantation for severe aplastic anemia. Bone Marrow Transplant. 12:225–231.

536

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 24 2001

28. Ikegami, M., T. R. Korfhagen, J. A. Whitsett, M. D. Bruno, S. E. Wert, K. Wada, and A. H. Jobe. 1998. Characteristics of surfactant from SP-A-deficient mice. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 275:L247–L254. 29. Spech, R. W., P. Wisniowski, D. L. Kachel, J. R. Wright, and W. J. Martin, II. 2000. Surfactant protein A prevents silica-mediated toxicity to rat alveolar macrophages. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 278:L713– L718. 30. Wang, J. Y., C. C. Shieh, P. F. You, H. Y. Lei, and K. B. Reid. 1998. Inhibitory effect of pulmonary surfactant proteins A and D on allergen-induced lymphocyte proliferation and histamine release in children with asthma. Am. J. Respir. Crit. Care Med. 158:510–518. 31. McIntosh, J. C., S. Mervin-Blake, E. Conner, and J. R. Wright. 1996. Surfactant protein A protects growing cells and reduces TNF-alpha activity from LPS-stimulated macrophages. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 271: L310–L319. 32. Schagat, T. L., M. J. Tino, and J. R. Wright. 1999. Regulation of protein phosphorylation and pathogen phagocytosis by surfactant protein A. Infec. Immun. 67:4693–4699. 33. Ohmer-Schrock, D., C. Schlatterer, H. Plattner, and J. Schlepper-Schafer. 1995. Lung surfactant protein A (SP-A) activates a phosphoinositide/cal-

cium signaling pathway in alveolar macrophages. J. Cell Sci. 108:3695–3702. 34. Oosting, R. S., M. M. van Greevenbroek, J. Verhoef, L. M. van Golde, and H. P. Haagsman. 1991. Structural and functional changes of surfactant protein A induced by ozone. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 261: L77–L83. 35. Blau, H., S. Riklis, J. F. Van Iwaarden, F. X. McCormack, and M. Kalina. 1997. Nitric oxide production by rat alveolar macrophages can be modulated in vitro by surfactant protein A. Am. J. Physiol. 272:L1198–L1204. 36. Kremlev, S. G., and D. S. Phelps. 1994. Surfactant protein A stimulation of inflammatory cytokine and immunoglobulin production. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 267:L712–L719. 37. Wright, J. R. 1997. Immunomodulatory functions of surfactant. Physiol. Rev. 77:931–962. 38. Wright, J. R., D. F. Zlogar, J. C. Taylor, T. M. Zlogar, and C. I. Restrepo. 1999. Effects of endotoxin on surfactant protein A and D stimulation of NO production by alveolar macrophages. Am. J. Physiol. (Lung Cell. Mol. Physiol.) 276:L650–L658. 39. McIntosh, J. C., A. H. Swyers, J. H. Fisher, and J. R. Wright. 1996. Surfactant proteins A and D increase in response to intratracheal lipopolysaccharide. Am. J. Respir. Cell Mol. Biol. 15:509–519.