Published ahead of print on October 24, 2003, doi:10.1165/rcmb.2003-0312OC
© 2004 American Thoracic Society DOI: 10.1165/rcmb.2003-0312OC Monocyte Recruitment into the Lungs in Pneumococcal PneumoniaMcDonald Research Laboratories and iCAPTURE Centre, University of British Columbia, St. Paul's Hospital, Vancouver, British Columbia, Canada Address correspondence to: Stephan F. van Eeden, M.D., Ph.D., McDonald Research Laboratory, University of British Columbia, St. Paul's Hospital, 1081 Burrard Street, Vancouver, BC, V6Z 1Y6 Canada. E-mail: svaneeden{at}mrl.ubc.ca
The recruitment of monocytes into the alveolar spaces is crucial for clearing infections and resolving the inflammatory response. We have previously reported the effect of acute pneumonia on monocyte transport through the bone marrow, and the present study concerns their clearance from the blood and migration into the lung airspaces. Dividing monocytes were labeled with the thymidine analog, 5'-bromo-2'-deoxyuridine (BrdU). Whole blood containing the labeled monocytes (MOBrdU) was transfused from either donor rabbits with pneumonia or from uninfected controls into recipients with pneumonia, where they were detected in blood and tissues using a double immunostaining method. The results show that MOBrdU from infected animals rapidly disappeared from the circulation (P < 0.05), preferentially sequestered in the infected lung tissue within 1 h (22.0 ± 3.3% versus 6.0 ± 0.4%, pneumonic region versus peripheral blood, P < 0.05), and accumulated to a greater degree in pneumonic airspaces than control monocytes 48 h after transfusion (3.9 ± 0.5% versus 1.1 ± 0.1%, P < 0.05). We conclude that immature monocytes released from the marrow by pneumonia sequester rapidly in lung microvessels but their migration in pneumonic airspaces is delayed.
Abbreviations: alkaline phosphatase and anti–alkaline phosphatase, APAAP • 5'-bromo-2'-deoxyuridine, BrdU • fluorescein isothiocyanate, FITC • monoclonal antibody, mAb • BrdU-labeled monocytes, MOBrdU • BrdU-labeled monocytes from pneumonic animal, MOBrdU(P) • BrdU-labeled monocytes from control animal, MOBrdU(C) • polymorphonuclear leukocytes, PMN • R-phycoerythrin, R-PE • total white blood cell, WBC
Effective host defense against infection of the lung depends on an orchestrated sequential recruitment of leukocytes (1). The polymorphonuclear leukocytes (PMN) migrate out of the pulmonary capillaries into the alveolar space within hours following Streptococcus pneumoniae instillation (2, 3). In contrast, monocyte recruitment occurs predominantly 24–48 h later (4–6). Monocytes have the capability to continue to divide in the lung and undergo maturation into macrophages with an enhanced ability to clear microorganisms when these pathogens are replicating and rapidly increasing in number (7). Thus a timely and adequate monocyte response is essential for controlling and clearing infection and terminating the inflammatory process. Disregulation of this response leads to ongoing infection, inflammation, widespread lung injury, and the clinical features of the adult respiratory distress syndrome (8, 9). Monocytes recruited into the lung originate from either an intravascular pool of monocytes (5) that are either circulating or marginating along the vessel walls, or from precursors in the bone marrow (6). Several laboratories (6, 10), including our own (11), have shown that pneumococcal pneumonia shortens monocyte transit time through the bone marrow and increases their release into the circulation. These newly released monocytes are immature because the maturation process in the marrow has been shortened, but they retain the ability to divide (6, 10, 11). Previous studies have shown that immature PMN marginate in the lung more quickly than the mature PMN but migrate into the tissues more slowly (2, 3). The present study examines the margination and migration of monocytes into the lung and shows that there are important differences from PMN. We used the thymidine analog, 5'-bromo-2'-deoxyuridine (BrdU) to label monocyte precursors in rabbits with or without pneumonia. The BrdU-labeled monocytes released into the circulation of these animals were collected in whole blood and transfused into recipient animals where their disappearance from the circulation, margination in lung microvessels, and migration into airspaces were studied using a double-labeling technique to identify the BrdU-labeled monocytes (11).
Animals Forty-nine female New Zealand white rabbits were used in this study. Nineteen (2.9 ± 0.4 kg, mean ± SEM) were used as blood donors and 30 (2.2 ± 0.3 kg) as recipients. Sedation (Fentanyl [20 µg/kg] and droperidol [1 mg/kg]) was administered by subcutaneous injection to facilitate blood collection or transfusion. The protocol was approved by the Animal Experimentation Committee of University of British Columbia.
Experimental Design Labeled monocytes harvested from the pneumonia animals (MOBrdU[P]) were assumed to include the monocytes released from the marrow by infection, whereas labeled monocytes from the control group (MOBrdU[C]) represented normal circulating monocytes. To verify these assumptions and determine the time course of appearance of MOBrdU in the circulation after the BrdU labeling, preliminary studies were performed on the infected animals (n = 3), the control animals (n = 3), and a third group (n = 3) in which 1 ml of saline (vehicle) was instilled in the lung before the BrdU labeling. Blood samples were obtained from the central ear artery of these rabbits before and at 8, 16, and 24 h after the first BrdU injection for immunohistochemical and flow cytometric assay described below. Sequestration and migration of MOBrdU. The MOBrdU recruited into the lung tissue were measured by transfusing the MOBrdU harvested from the donors into the recipient rabbits in whole blood using a method previously described in detail (11, 13). Immediately before the transfusion of MOBrdU, a focal pneumonia was induced in the recipients as follows. The recipient rabbits were anesthetized as described above. S. pneumoniae (1.5 x 109 organisms) were mixed in 1 ml of sterile saline and instilled into the lower lobe of the lung as described above. An equal volume of sterile saline was instilled into the contralateral lower lobe of the same rabbit that served as an internal control (control region). Animals were killed at 1, 8, and 48 h (n = 6, each time point) after transfusion with an overdose of sodium pentobarbital. The chest was opened rapidly and the base of the heart was ligated to maintain the pulmonary blood volume. These time points were selected because (i) 1 h is needed for infused leukocytes to reach a steady state in the circulation and sequestrate between the circulating and marginating pool (11, 13), (ii) the 8 h time point represents PMN recruitment into the airspace (2, 3), and (iii) the 48 h time point represents a switch from PMN to predominantly mononuclear cell recruited into airspaces (6, 7, 14). The trachea and both lungs were separated from other organs and inflated by intratracheal instillation of OCT compound (Miles Laboratories, Elkhart, IN). Then the volume of each lung was measured by the water displacement. The lungs were frozen in liquid nitrogen and stored at –80°C until used. Clearance of MOBrdU from the circulation. The behavior of transfused MOBrdU in the recipient circulation of rabbits with pneumonia was measured from blood samples taken from the central ear artery of the recipient rabbits before and at 1, 8, 24, and 48 h after the blood transfusion. One milliliter of blood was collected in standard Vacutainer tubes containing EDTA (Becton Dikinson, Rutherford, NJ). Total white blood cell (WBC) counts were determined on a model SS80 Coulter Counter (Coulter Electronics, Hilaeh, FL). Differential counts of monocytes were obtained by counting 200 leukocytes in randomly selected fields of view on Wright-Giemsa stained blood smears. Using these values, the number of MOBrdU in the circulation of each recipient was calculated and expressed as a fraction of the total number of labeled monocytes originally infused and corrected for the calculated blood volume (15) of the recipient as previously described (11).
Immunohistochemical Detection of MOBrdU Double immunoenzymatic staining of monocytes/macrophages for RbM2 and BrdU. Cytospins prepared from the blood samples (11, 13) were used to identify the transfused MOBrdU using the alkaline phosphatase and anti–alkaline phosphatase (APAAP) method (16) and a double immunolabeling technique that has been fully described elsewhere (11). Briefly, the monocytes/macrophages were identified by RbM2 (ICN Biomedicals, Aurora, OH) (17) and the anti-BrdU mAb Bu20a (DAKO Laboratories, Copenhagen, Denmark) was used to identify the transfused cells and determine the fraction of MOBrdU. Cytospin slides were coded and MOBrdU were evaluated (11) by investigators without knowledge of the treatment protocol. The lung sections (6–8 µm thick) were incubated with 5% rabbit serum for 15 min before application of 0.5 µg/ml of RbM2 for 90 min in a humidity chamber at room temperature. Nonspecific mouse IgG1 at 0.5 µg/ml was used as a negative control. A 1:20 dilution of rabbit antimouse IgG (DAKO) was applied and followed by the antimouse IgG alkaline phosphatase–conjugated complex (DAKO) in a 1:50 dilution for 20 min, respectively. The alkaline phosphatase was developed for 20 min in a new fuschin-based, red substrate solution to detect specific bindings. Specimens were then fixed with 1% paraformaldehyde for 15 min before staining for the presence of nuclear BrdU using a second APAAP procedure as described in detail previously (11, 18, 19). The alkaline phosphatase was developed with a commercially available kit, HistoMark Blue (Kirkegaard and Perry, Gaithersburg, MD), for 10 min in the dark.
Morphometric Studies of Lung Tissue A point-counting grid was placed over the slide of the whole lung slices stained with hematoxylin and eosin, and examined at x4 magnification (Level 1). The number of points falling in the parenchyma of each lower lobe was expressed as a fraction of the total number of points falling in the entire lung to determine the volume fraction of the pneumonic or saline-treated lung. The number of points falling in the upper lobe divided by the total points in the lung provided the volume fraction of the untreated lung.
The histologic sections (6–8 µm thick) cut from the samples of pneumonic, control, and untreated sites were used to capture 10 randomly selected images from each area using a spot digital camera (Microspot; Nikon Inc., Tokyo, Japan) at x400 magnification. The images were coded and examined without knowledge of the group. Those images were analyzed using a point counting grid of 567 (21 x 27) points that was superimposed onto the captured image and image analysis software (Image Pro Plus; Media Cybernetics, Silver Spring, MD). They were evaluated in each lung region (Level 2) to determine the volume fraction of airspace and tissue and the volume fraction of the airspace and tissue occupied by MOBrdU or monocytes/macrophages. The results are expressed as the fraction of monocytes/macrophages in airspace or tissue that are BrdU-labeled. The number of MOBrdU or monocytes/macrophages in the airspace or tissue from the pneumonic, saline-treated region and untreated region was calculated separately as an example:
Flow Cytometric Analysis of Circulating Monocytes
Statistical Analysis
S. pneumoniae instillation caused a more rapid release of monocytes from the marrow compared with the noninstilled control animals (Figure 1A), especially at 16 h (76 ± 3.9% versus 51 ± 5.3%, MOBrdU[P] versus MOBrdU[C], P < 0.05). The circulating monocytes in the pneumonic donor animals also expressed higher levels of L-selectin on their surface (a marker of newly released leukocytes) (Figure 1B) and similar levels of CD11b (activation marker) (Figure 1C) following the instillation of bacteria. In contrast, no difference was seen in the release of monocytes from the marrow (Figure 1A) or the expression of surface marker on the circulating monocytes (data not shown) between animals instilled with vehicle only and with no instillation (control, MOBrdU[C]).
Double Immunolabeling of Monocytes/Macrophages The staining of circulating monocytes in blood and monocytes/macrophages in the lung for nuclear BrdU (blue) and cytoplasm RbM2 (red) are shown in Figures 2A and 2B, respectively. All RbM2 positive cells (monocytes or macrophages) were classified as either BrdU-positive or -negative. The expression of both RbM2 and BrdU was similar in paired single- and double-stained slides (data not shown).
Clearance of MOBrdU from the Circulation Figure 3 shows the clearance of MOBrdU from the circulation in recipients. Following the transfusion of donor whole blood containing labeled leukocytes, the time required to achieve a steady state in the circulation is  1 h (11, 13). The fraction of MOBrdU(P) in the circulation at 1 h after transfusion (37 ± 6.7%) was similar compared with MOBrdU(C) (41 ± 7.6%). Labeled monocytes obtained from animals with pneumonia, MOBrdU(P) cleared from the circulation more rapidly compared with control monocytes, MOBrdU(C), transfused from animals without pneumonia (at 8 h; P < 0.05).
Sequestration and Migration of Monocytes Figure 4 shows the number of monocytes/macrophages accumulated in the lung tissue (A) and airspaces (B) in S. pneumoniae–treated (pneumonic) and saline-treated (control) regions of the lung at different time points after the instillation of bacteria. There were more monocytes/macrophages in pneumonic tissue (alveolar walls) and airspaces than in the contralateral control tissue or airspaces (at 8 h in the tissue, P < 0.05; at 48 h in the tissue and airspace, P < 0.01). Instillation of saline (vehicle) did not influence total number of monocyte/macrophages that accumulated or migrated (ratio 1) (Figures 4A and 4B).
Figure 5 shows the fold increase of labeled monocytes (MOBrdU) in the pneumonic and control recipient lungs. No changes were seen in the number of MOBrdU in the untreated region over the study period (data not shown). The labeled monocytes from donor animals with pneumonia, MOBrdU(P), rapidly accumulated in both the pneumonic region and the vehicle instilled control region of the recipient animals compared with labeled monocytes from control animals, MOBrdU(C) (P < 0.05; at 1 and 8 h in the pneumonic tissue [A] and at 48 h in the control tissue [B]). But this difference between MOBrdU(P) and MOBrdU(C) groups in the pneumonic lung disappeared by 48 h when the number of MOBrdU(C) accumulated in the pneumonic region was higher compared with that of control region (the ratio at 48 h was 5.2 ± 0.9 to 1.7 ± 0.4, P < 0.05).
Figure 6 shows the number of MOBrdU in each group migrated into airspaces of the recipients. There was no difference in migration of MOBrdU in any lung region in the first 8 h. But the migration of the labeled monocytes from animals with pneumonia (MOBrdU[P]) into airspaces of the pneumonic region of the recipients tended to be higher than labeled monocytes from control animals (MOBrdU[C]) at 8 and 48 h (P < 0.05 at 48 h, Figure 6A).
Figure 7A shows the fraction (%) of labeled monocytes (MOBrdU) in lung tissues (alveolar walls) compared with circulating blood 1 h after the transfusion (steady state). The fraction of the MOBrdU(P) in pneumonic tissue was higher than those of control or untreated tissue and also higher than that in peripheral blood (P < 0.05). The fraction of the MOBrdU(C) was similar in all lung tissues and peripheral blood. Moreover, Figure 7B shows the fraction of MOBrdU in the pneumonic tissues and pneumonic airspace at 48 h after the transfusion. The fraction (%) of labeled monocytes from animals with pneumonia migrated in the pneumonic region was higher than migration of labeled monocytes from control animals (3.9 ± 0.5% versus 1.1 ± 0.1%, MOBrdU[P] versus MOBrdU[C], P < 0.05), indicating preferential migration of MOBrdU(P).
We have recently reported that the monocytes appear in the circulation much earlier than PMN following pulse labeling their precursors in the bone marrow (11, 12, 24), and that an acute focal pneumonia accelerates this process by shortening their transit through the marrow pools (11). The present study used a combination of donor and recipient animals to extend these findings by showing that the immature monocytes released into the circulation by pneumonia are more rapidly cleared from the circulation and preferentially recruited to the airspaces of the infected lung tissue. To avoid the problem of labeled monocytes being added by release from the marrow or division in the lung, we measured the clearance of monocytes from the blood and their recruitment into infected tissue by transferring BrdU-labeled cells present in the blood of donor animals into recipients that had not received BrdU injection. The techniques are fully described elsewhere (11, 13, 18) and were previously used to show that the peripheral blood monocyte population of the animals with pneumonia is enriched with cells that have been rapidly released from the marrow (11). The present results suggest that these immature monocytes are primed for recruitment into the airspace of the infected lung tissue. The observation that labeled monocytes appeared more rapidly in the circulation of the donor animals with pneumonia (Figure 1A) is consistent with our previous report that pneumonia accelerates monocyte turnover in the marrow (11). The high level of L-selectin on these circulating monocytes from donor animals with pneumonia compared with the noninfected donors (Figure 1B) is also consistent with the rapid release of new monocytes from the marrow (2, 25). Furthermore, the similarity in the expression of the surface cell activation marker CD11b (Figure 1C) indicates that monocyte margination was not increased by a mechanism that involves the CD11b system such as cell activation. Under steady-state conditions, mature monocytes circulate with an average half-life of 12–24 h before they leave the vascular space to enter the tissues (11, 14, 26). The present results show that monocytes transferred from donor animals with pneumonia disappear from the circulation of recipients with pneumonia more rapidly than monocytes from noninfected controls (Figure 3). This behavior is similar to that previously reported for PMN (12), and shows that both of these cell types marginate more quickly during pneumococcal pneumonia than in the control state. But our results also show that the subsequent recruitment of these two cell types into the lung airspace is quite different. Studies of the temporal relationship of PMN and monocyte recruitment into an inflammatory site in the lung (6, 7, 14) show that PMN begin to accumulate in the airspaces of the lung within 1 or 2 h after the initiation of infection, compared with the 24–48 h delay for the monocytes. Our results showing that monocytes accumulate in the alveolar walls and airspaces of the pneumonic regions of the lung between 8 and 48 h after the initiation of the infection (Figures 4A and 4B) are consistent with these previous reports. They also extend this observation by showing that pneumonia releases a population of immature BrdU-labeled monocytes that accumulate more rapidly in the alveolar wall tissue of both the infected (Figure 5A) and noninfected lung (Figure 5B) of recipient animals with pneumonia. The cells from the pneumonic donors also appear in the airspaces of the infected region of the recipient lung to a greater degree than the cells transferred from animals without pneumonia (Figure 6). Furthermore, Figure 7A shows that 1 h after the transfusion of the labeled monocytes from the pneumonic animals, they had accumulated to a greater degree in the infected region of the recipient lung compared with the sham-infected and untreated control regions. And Figure 7B shows that at 48 h there was a greater accumulation of the labeled cells transferred from the infected animals into the infected recipient airspaces. These results strongly suggest that monocytes released from the marrow during pneumococcal pneumonia are conditioned for preferential recruitment into the pneumonic lung. This behavior of the monocytes is similar to PMN in that both cell types marginate to a greater degree in infected regions of the lung (2, 3). The widespread margination of PMN released into the circulation during a pneumonia has been attributed to their size and a decrease in their deformability (27, 28). The role of PMN adhesion molecules in their ability to marginate are unclear, but Doyle and colleagues have shown that L-selectin does not play a role in the immediate sequestration of PMN in lung capillaries but does contribute to prolonged margination (29). Similar mechanisms could explain the prolonged margination of labeled monocytes because they also have an elevated L-selectin expression on their surface (Figure 1B) and are probably less deformable than more mature cells. Furthermore, Lundahl and Hed (4) demonstrated that shedding of L-selectin on monocytes is less pronounced as compared with PMN, suggesting that sustained expression of L-selectin on newly released monocytes could contribute to their prolonged margination in lung microvessels. These possible mechanisms of monocyte margination during pneumonia need further investigations. The delayed appearance of monocytes in the air spaces compared with PMN has been attributed to a requirement for prior accumulation of PMN in the area of inflammation to condition the tissue for monocyte migration (1, 14, 30). Others (4) have suggested that the greater expression of L-selectin on these cells slows the migratory process. The present data extend these possibilities by showing that the monocytes released into the circulation by pneumonia marginate in the microvessels of the infected lung tissue within an hour and then appear in the airspaces in greater numbers than monocytes from uninfected controls by 48 h. As the protocol of the present studies did nothing to interfere with conditions at the site of infection, the difference in behavior between monocytes transferred from animals with pneumonia and those transferred from uninfected controls strongly suggests that a monocyte population released from the marrow by pneumonia is conditioned for recruitment into the site of the infection. This underscores the participation of monocytes newly released from the bone marrow as a component of lung defense during lung infection. Because these immature monocytes also have a greater potential to divide than those released under normal conditions, we speculate that the prolonged time required for the appearance of this monocyte population in the airspaces may involve the time required for the cells to divide and increase the numbers of migrating monocytes.
The authors gratefully thank Amrit Samra, Cristin Collier, Diane Minshall, and Yukio Sato for technical support. This work was supported by a grant from the BC Lung Association, Canadian Institute for Health Research (grant #4219). S.v.E. is the recipient of a Career Investigators award from the American Lung Association and the William Thurlbeck Distinguished Researcher Award. Received in original form August 20, 2003 Received in final form October 20, 2003
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
P < 0.01). The number of macrophages increased in airspaces in the pneumonic region just at 48 h (
