Introduction

Nitric oxide (NO) generated by NO synthase (NOS) plays an important role in physiologic and pathophysiologic processes in the lung including vasotone and bronchotone regulation, neurotransmission, immune defense, and mucin secretion (Gaston et al, 1994; Moncada et al, 1991). Three isoenzymes of NOS have been described (Gaston et al, 1994; Moncada and Higgs, 1993): the neuronal isoform bNOS (NOS-1) and an endothelial isoform eNOS (NOS-3) as constitutive isoenzymes, and a mitogen-inducible isoform iNOS (NOS-2). Cellular expression of bNOS and eNOS in different cell types of normal lung tissue and pulmonary cell cultures has been reported earlier (Kobzik et al, 1993; Shaul et al, 1994; Sherman et al, 1999). Cellular expression of iNOS in the lung parenchyma has so far only been observed after experimental endotoxin stimulation or in sections from patients suffering from inflammatory lung disease (Kobzik et al, 1993; Robbins et al, 1994; Tracey et al, 1994). However, recently all three NOS isoforms were shown by immunohistochemistry to be expressed in the developing lung, and all were assumed to contribute to NO generation functioning in normal lung physiology (Sherman et al, 1999; Xue et al, 1996). A recent study performed in human airway tissue originating from patients with bronchial carcinoma undergoing lung resection showed iNOS expression in bronchial epithelial cells (Watkins et al, 1997).

Extensive production of NO has been assumed to be a key event in the pathogenesis of several inflammatory lung diseases, such as sepsis-induced lung injury and asthma (Hamid et al, 1993; Hinder et al, 1999; Kristof et al, 1998). The inducible isoform iNOS can be activated by bacterial lipopolysaccharides (LPS, endotoxin) and cytokines (Kristof et al, 1998; Nakayama et al, 1992; Robbins et al, 1994) and—via excess NO formation—may cause inappropriate vasodilation, which is a key feature in septic organ failure (Stewart et al, 1995; Ullrich et al, 1999). In addition, iNOS-dependent excessive NO generation may induce tissue damage through the formation of peroxynitrite, and such an event has been implicated in the pathogenesis of LPS-induced lung injury (Kooy et al, 1995; van der Vliet et al, 1999; Wizemann et al, 1994). Studies in mice deficient in iNOS gene activity were shown to be more resistant to LPS-induced acute lung injury than wild-type animals (Kristof et al, 1998; Weimann et al, 1999).

Because of the vasodilatory effects of NO, inhalation of this agent has been used as a therapeutic approach to decrease the pulmonary vascular resistance in well-ventilated, ie, NO-accessible lung areas, and thereby improve oxygenation in patients with acute respiratory distress syndrome (ARDS) (Michael et al, 1998; Payen, 1998; Troncy et al, 1998). Unfortunately, NO inhalation, although initially improving gas exchange, did not lead to sustained improvement of oxygenation compared with conventional therapy in the hitherto performed controlled studies (Michael et al, 1998; Payen, 1998). Nonvasodilation-related effects of inhaled NO, currently not well understood, have been suggested to explain this finding. Moreover, too little information exists about the regulation of the different NOS isoenzymes, and thus the background of endogenous NO formation in ARDS and septic lung failure is still unclear (Payen, 1998). Such regulation may apparently include iNOS and enhanced expression of iNOS in several cell types such as epithelial cells (Kobzik et al, 1993; Warner et al, 1995; Watkins et al, 1997), alveolar macrophages (Kobzik et al, 1993; Tracey et al, 1994), or endothelial cells (Kobzik et al, 1993; MacNaul and Hutchinson, 1993), which has been reported in several studies in response to mitogen stimulation. In addition, regulation of eNOS and bNOS may be affected, however, only minimal information about the regulation of these constitutive isoforms in response to LPS or cytokines currently exists. Some recent evidence suggests that the activity of the constitutive eNOS may be down-regulated during inflammation (MacNaul and Hutchinson, 1993; Schwartz et al, 1997).

In the present study, we assessed cell type–specific regulation of the different NOS isoenzymes in rat lung tissue in response to intravascularly applied LPS by a previously described method of in situ microdensitometry (Ermert et al, 2000a, 2001). In a recent study using this technique, cyclooxygenase-2 (Cox-2) was shown to be time-dependently and LPS dose-dependently up-regulated in different cells of isolated rat lungs challenged with endotoxin (Ermert et al, 2000b). Interestingly, several cell types such as vascular smooth muscle cells (VSMC) of large arteries, endothelial cells, or bronchial smooth muscle cells displayed such Cox-2 up-regulation only in the presence of plasma constituents in the buffer perfusate. This plasma dependence was correlated with the lack of expression of membrane CD14 in these cell types, again demonstrated by immunohistochemistry (Ermert et al, 2000b). Several effects of LPS are known to be mediated by the plasma protein LBP (LPS-binding protein) and the membrane receptor CD14 (Pugin et al, 1993; Tobias et al, 1989; Ulevitch, 1993). Moreover, a soluble form of CD14 has been described that forms complexes with LPS and LBP and may then bind to cells negative for membrane CD14 (Frey et al, 1992; Pugin et al, 1993). Recently, additional LPS receptors (CD55, Toll-like receptors) have been detected, which may serve as coreceptor(s) for CD14 providing an intracellular domain for signal transduction, because CD14 is void of such a domain (El Samalouti et al, 1999; Qureshi et al, 1999; Yang et al, 1998).

Against this background, the current investigation again used perfused rat lungs, stimulated with LPS in the absence or presence of plasma constituents, to address the profile of lung cellular NOS expression under baseline conditions and in response to microbial challenge.

Results

Immunohistochemistry of NOS isoenzymes shows constitutive expression of all three NOS isoenzymes in rat lung tissue (Fig. 1). Staining was absent in sections where the primary antibody was omitted or nonspecific immune serum was used (Fig. 1D). Background staining was low (mean gray value <100). Nerve fibers, which were positive for all three isoenzymes (Tables 1a, 1b, 2a, 2b, 3a and 3b), were not measured by image analysis because they were not regularly present in all lung sections.

Figure 1
figure 1

Immunohistochemical localization of neuronal/brain nitric oxide synthase (bNOS) (A), inducible (i)NOS (B) and endothelial (e)NOS (C) in normal rat lung tissue. (A). Strong bNOS immunostaining was detected in bronchial epithelial cells of small bronchioli within the alveolar tissue. (B) Constitutive iNOS expression was found in bronchial epithelium and bronchial smooth muscle cells. Vascular smooth muscle cells (VSMCs) and especially endothelial cells of pulmonary vessels showed positive immunostaining for eNOS (C). Control staining with nonspecific immune serum shows no staining reaction (D).

Full size image
Table 1a Quantification of eNOS Staining Intensity—Experiments in the Absence of Plasma (Mean Gray Values)
Full size table
Table 1b Quantification of eNOS Staining Intensity—Experiments in the Presence of Plasma (Mean Gray Values)
Full size table
Table 2a Quantification of iNOS Staining Intensity—Experiments in the Absence of Plasma (Mean Gray Values)
Full size table
Table 2b Quantification of iNOS Staining Intensity—Experiments in the Presence of Plasma (Mean Gray Values)
Full size table
Table 3a Quantification of bNOS Staining Intensity—Experiments in the Absence of Plasma (Mean Gray Values)
Full size table
Table 3b Quantification of bNOS Staining Intensity—Experiments in the Presence of Plasma (Mean Gray Value)
Full size table

eNOS

Control Lungs.

In normal rat lung tissue, bronchial epithelial cells exhibited a strong eNOS immunostaining, which could be detected in all large and small bronchi (Table 1, a and 1b; Fig. 2A). Similarly, endothelial cells were strongly stained (Fig. 2, C and E). In addition bronchial smooth muscle cells (Fig. 2A) and VSMCs of fully muscular and partially muscular vessels (Figs. 1C and 2E) showed marked immunoreactivity. Smooth muscle cells of large arteries at the hilum and myocytes of large hilar veins were not stained (Table 1, a and 1b; Fig. 2C). In addition, peribronchial and perivascular nerve fibers, cells within the bronchus-associated lymphoid tissue (BALT), and alveolar macrophages exhibited positive staining for eNOS (Table 1, a and 1b). No difference between controls with buffer-perfusion and those with buffer/plasma-perfusion was noted.

Figure 2
figure 2

Image analysis with pseudocolor depiction of immunohistochemical eNOS staining in control (A, C, and E) and lipopolysaccharide (LPS)-stimulated rat lung tissue (B, D, and F). eNOS staining intensity was decreased in bronchial epithelial cells in response to LPS, whereas the expression of this isoenzyme in bronchial smooth muscle cells remained unchanged (B, 1000 ng/ml LPS, 2 hours). Likewise, strong endothelial eNOS expression (C) was found to decrease after LPS challenge (D, 10,000 ng/ml LPS, 2 hours). Note that muscle cells of the artery wall (C and D) are not stained. Staining intensity of VSMC and endothelium of small muscular vessels (E) is markedly decreased in LPS-treated lungs (F, 10,000 ng/ml LPS, 2 hours).

Full size image

Expression of eNOS after LPS.

In response to LPS, a significant decrease in eNOS staining intensity was detected within 1 hour in nearly all cell types that expressed eNOS under baseline conditions. Decrease in staining intensity is visualized by pseudocolor depiction, eg, in bronchial epithelial cells of large bronchi (Fig. 2, A and B), endothelial cells (Fig. 2, C and D), or VSMCs of small muscular vessels (Fig. 2, E and F). The decrease in staining intensity occurred within 1 hour and was further enhanced within 2 hours of LPS incubation (Table 1, a and 1b; Fig. 4). In addition, the LPS-induced decrease in eNOS staining intensity was dose-dependent and already evident upon low-dose LPS challenge (50 ng/ml LPS), in particular in bronchial epithelial cells, alveolar macrophages, and VSMCs (Table 1, a and 1b).

Figure 4
figure 3

Quantitative evaluation of immunohistochemical eNOS staining in response to LPS. Staining intensity is depicted in percent of gray scale values measured, with control lungs being set at 100% (except for D, where no control cell staining was noted). LPS challenge was performed in the absence (open bars) and in the presence (closed bars) of 1.5% plasma in the buffer perfusate. Time- and dose-dependent down-regulation (A to C) was noted in most cell types. In contrast increased expression of eNOS was found in mast cells (D). Mean ± sem of five independent experiments each are given. *p < 0.05, **p < 0.01, ***p < 0.001, as compared with control lung cells.

Full size image

Staining intensity was not found to be decreased after LPS application in bronchial smooth muscle cells (Table 1, a and 1b; Fig. 2, A and B). Immunostaining of eNOS was newly detected in mast cells located in the perivascular and peribronchial connective tissue (Table 1, a and 1b; Fig. 4). Moreover, in mast cells a significant increase of eNOS expression was noted in groups exposed to 10,000 ng/ml LPS for 2 hours as compared with 1 hour. There was no difference in the cellular eNOS expression and regulation pattern in groups perfused in the absence (Table 1a) versus the presence (Table 1b) of 1.5% plasma.

iNOS

Control Lungs.

Strong iNOS immunoreactivity was localized to bronchial epithelial cells mainly of large first- and second-generation bronchi within normal rat lung tissue (Table 2, a and 2b; Figs. 1B and 3A). In addition, cells of the BALT, single cells within the alveolar septum, and alveolar macrophages (Fig. 3C) showed positive immunoreactivity. Bronchial smooth muscle cells (Figs. 1B and 3A), smooth muscle cells of fully muscular and partially muscular vessels, and myocytes of large hilar veins (Fig. 3E) displayed moderate iNOS immunostaining (Table 2, a and 2b) in contrast to large arteries, which were not stained. Perivascular leukocytes, mast cells in the perivascular and peribronchial connective tissue, and endothelial cells were not stained in control lungs. In accordance with eNOS and bNOS immunostaining, no difference between controls with buffer-perfusion and those with buffer/plasma-perfusion was noted.

Figure 3
figure 4

Image analysis with pseudocolor depiction of immunohistochemical iNOS staining in control (A, C, and E) and LPS-stimulated rat lung tissue (B, D, and F). iNOS staining intensity was increased in bronchial epithelial cells and bronchial smooth muscle cells in response to LPS (B, 1000 ng/ml LPS, 1 hour). Within the lung parenchyma, alveolar macrophages and cells in the alveolar septum showed marked up-regulation of iNOS within 1 hour after LPS application (D, 10000 ng/ml LPS, 1 hour). Cardiac myocytes of large hilar veins display extensively increased iNOS expression in LPS-challenged lungs (F, 1000 ng/ml LPS, 2 hours).

Full size image

Expression of iNOS after LPS.

After LPS stimulation, iNOS expression was increased in all cell types that showed iNOS immunoreactivity in control lungs (Table 2, a and b). The increase in staining intensity of iNOS displayed time and dose dependency (Fig. 5). Even challenge with low-dose LPS (50 ng/ml) sufficed to provoke some increase in staining intensity, which was, however, not statistically significant in most cell types.

Figure 5
figure 5

Quantitative evaluation of immunohistochemical iNOS staining in response to LPS. Staining intensity is depicted in percent of gray scale values with control lungs being set at 100%. LPS challenge was performed in the absence (open bars) and in the presence (closed bars) of 1.5% plasma in the buffer perfusate. Time- and dose-dependent up-regulation (A to C) was noted in most cell types. Mean ± sem of five independent experiments each are given. *p < 0.05, **p < 0.01, ***p < 0.001, as compared with control lung cells.

Full size image

Bronchial epithelial cells of large bronchi (Fig. 3, A and B) and myocytes of large hilar veins (Fig. 3, E and F) displayed the highest staining intensities under challenge with LPS as compared with other cell types. In addition, an increase in staining intensity was also evident in bronchial smooth muscle cells (Figs. 3, A and B, and 5) and alveolar macrophages (Figs. 3, C and D, and 5). In VSMC of fully muscular and partially muscular vessels, iNOS expression was also increased (Table 2, a and 2b). In endothelial cells, newly induced iNOS was detected in groups receiving 50 ng/ml LPS both in groups perfused with buffer alone or buffer/plasma. Staining of endothelial cells was further increased in all groups receiving 1,000 or 10,000 ng/ml LPS. Perivascular leukocytes, which did not display positive iNOS staining in control groups or groups receiving low-dose LPS showed time-dependent up-regulation of iNOS in response to 1,000 and 10,000 ng/ml LPS (Table 2, a and 2b). There was no significant difference noted between buffer (Table 2a) and buffer/plasma (Table 2b) perfused lungs.

bNOS

Control Lungs and LPS-Challenged Lungs.

In buffer- or buffer/plasma-perfused control lungs, strong bNOS immunostaining (Table 3, a and 3b) was detected in the bronchial epithelial cells of all bronchi, with strongest expression found in the smaller bronchi (Fig. 1A). In addition bNOS was present in perivascularly and peribronchially located mast cells. Next to alveolar macrophages, cells within the BALT exhibited positive immunostaining for bNOS. In addition in endothelial cells and VSMC of fully muscular vessels and myocytes of large veins at the lung hilum, a moderate expression of bNOS was detected. Nerve fibers exhibited intense bNOS immunoreactivity.

LPS treatment for either 1 or 2 hours, under both buffer- and buffer/plasma-perfusion conditions, did not change the pattern and intensity of bNOS immunostaining as compared with the control groups; data of selected experimental groups (1,000 ng/ml LPS, 2 hours; 10,000 ng/ml LPS, 1 hour and 2 hours) are presented in Table 3, a and 3b.

Discussion

The present study shows differential expression of all three types of isoenzymes in normal rat lung tissue. This was anticipated for the constitutive isoforms bNOS and eNOS (Asano et al, 1994; Kobzik et al, 1993; Shaul et al, 1994; Steudel et al, 1999), although an entire profile of cellular localization of both isoforms in lung tissue has so far not been shown. A recent study (Steudel et al, 1999) reported eNOS expression in epithelial cells, endothelial cells, and VSMC in normal rat lung tissue, which corresponds well with the data presented in this study. Alveolar macrophages and bronchial smooth muscle cells were, however, found to be negative for eNOS under baseline conditions in that study in contrast to the present investigation, which may be a result of different tissue treatment and embedding techniques (Steudel et al, 1999). Bronchial epithelial cells, endothelial cells, and VSMC of fully muscular vessels showed the highest staining intensity for eNOS under baseline conditions, as quantified by the currently used in situ microdensitometric method (Ermert et al, 2001). bNOS expression has been previously localized to nerve, epithelial, and endothelial cells in rat and human lung tissue (Asano et al, 1994; Kobzik et al, 1993). In addition we report here bNOS expression for alveolar macrophages and VSMC of fully muscular vessels, as well as myocytes of hilar veins. Perivascular and peribronchial macrophage-like cells, which in previous studies were shown to represent mast cells to some major extent (Ermert et al, 1998, 2000b), were shown to express bNOS but not eNOS or iNOS in nonstimulated rat lung tissue.

Similar to eNOS and bNOS, iNOS was found to be constitutively expressed in several lung cell types under baseline conditions in the present study. In accordance with previous observations (Asano et al, 1994; Steudel et al, 1999; Tracey et al, 1994; Watkins et al, 1997), iNOS expression was noted in bronchial epithelial cells of large and small airways. Such prominent constitutive iNOS expression in the bronchial epithelium may reflect the fact that this epithelium represents a barrier with a large external surface, being naturally exposed to airborne noxious and microbial agents even in the absence of lung infection. In addition, iNOS was expressed in bronchial smooth muscle cells, myocytes of pulmonary hilar veins (this study; Steudel et al, 1999), alveolar macrophages, cells within the BALT, and VSMC of partially and fully muscular vessels (this study). Microdensitometry of immunohistochemical staining intensity showed the most intensive iNOS expression in control lungs to be present in bronchial epithelial cells of large bronchi and in myocytes of large hilar veins. Endothelial cells were found to express eNOS and bNOS but not iNOS under baseline conditions. Differential expression of NOS isoenzymes was also found in myocytes of large hilar veins, which expressed iNOS and bNOS but not eNOS in control lungs. A complex cellular pattern of NO generation via different isoenzymes thus seems to contribute to NO-dependent physiologic regulations, which were suggested to include those of vascular tone, bronchial tone, neurotransmission, or immune defense (Gaston et al, 1994; Moncada et al, 1991).

Excess NO production, causing inappropriate vasodilation and peroxynitrite-related cellular injury, has been implicated in the pathogenesis of ARDS and septic lung failure (Beckman and Koppenol, 1996; van der Vliet et al, 1999). Such NO overproduction was particularly attributed to the induction of iNOS in the lung tissue (Fujii et al, 1998; Warner et al, 1995; Wizemann et al, 1994). The current study does, however, demonstrate that the scenario is much more complex, with evidence being forwarded for a distinct regulation of both iNOS and eNOS in various lung cell types in response to endotoxin. eNOS expression was found to be down-regulated in all lung compartments, in nearly all cell types expressing this isoenzyme under baseline conditions. This was a striking finding of the present investigation, with underlying mechanisms for this type of regulatory response to endotoxin being currently unknown. In previous studies in cultured cells, a down-regulation of eNOS, triggered by increased NO production via LPS-induced iNOS expression, has been reported (Buga et al, 1993; MacNaul and Hutchinson, 1993; Schwartz et al, 1997). To our knowledge, such negative feedback regulation of eNOS expression by enhanced iNOS-derived NO formation has so far not been demonstrated in whole lung tissue but may well be operative to explain the disparate regulations of iNOS and eNOS in the present study. However, the down-regulation of eNOS in the vast majority of cell types in LPS-stimulated lungs may, of course, also be fully independent of enhanced iNOS-derived NO synthesis but may represent a direct downstream signaling event in endotoxin-exposed pulmonary cells. The underlying mechanism thus obviously deserves further elucidation.

Interestingly, de novo expression of eNOS was noted in mast cells in response to LPS. These cells were shown to express bNOS but not eNOS or iNOS in control lung tissue. The present observation is basically in line with earlier findings of TNF-mediated enhanced production of NO in cultured peritoneal and intestinal mast cells (Bissonnette et al, 1991), however, NO production was at that time not attributable to a certain NOS isoform. There is inconsistent information about NOS isoenzyme expression in mast cells of different organs: positive bNOS immunoreactivity has been reported for human skin mast cells (Shimizu et al, 1997), whereas both mouse myometrial (Huang et al, 1995) and connective tissue mast cells (Bidri et al, 1997) showed iNOS expression after hormonal or antigen stimulation. Further studies should address the issue of whether the up-regulation of eNOS in response to LPS in mast cells as opposed to other cell types is a lung-specific finding or is representative for mast cells of other organ origin.

In contrast to iNOS and eNOS, the expression of bNOS was found not to be altered in any of the different cell types in response to LPS stimulation. bNOS expression thus seems to be regulated in an entirely independent fashion, not comparable to that of eNOS and iNOS. Moreover, it does not seem to be affected by an increase in endogenous NO production, as expected because of the marked iNOS up-regulation, which is in line with previous in vitro studies (Buga et al, 1993; MacNaul and Hutchinson, 1993; Schwartz et al, 1997).

LPS exposure evoked increased expression of iNOS in particular in bronchial epithelial cells, bronchial smooth muscle cells, cells of the BALT, and alveolar macrophages. In the vascular system, iNOS was up-regulated in VSMC of fully muscular and partially muscular vessels and in myocytes of hilar veins. De novo expression in response to LPS, in a time- and dose-dependent fashion, was noted in endothelial cells and perivascularly located leukocytes. Increased expression of iNOS in leukocytes may well correspond to enhanced proinflammatory and cytotoxic activity of leukocytes during LPS-induced lung inflammation (Fujii et al, 1998; Liu et al, 1997). iNOS expression in pulmonary macrophages has been implicated in the regulation of the innate immunity system, which is indeed expected to respond to endotoxin challenge (Fujii et al, 1998; Liu et al, 1997). In cultured epithelial cells, increased production of NO has been observed to occur 8 to 24 hours after mitogen challenge and was associated with elevated iNOS mRNA and protein (Asano et al, 1994; Robbins et al, 1994; Watkins et al, 1997). In intact lungs, in contrast, increased concentrations of exhaled NO were reported as early as 2 to 3 hours after LPS treatment and were suggested to be an early marker of lung inflammation (Stewart et al, 1995). This study demonstrates significant increase in iNOS protein expression in bronchial epithelial cells within 1 to 2 hours after intravascular LPS application, which supports the notion that iNOS in particular contributes to excessive inflammatory NO generation in epithelial cells, and this is a very rapid response. Increase of iNOS in bronchial epithelium and alveolar macrophages has been correlated with enhanced nitrotyrosine detection, a marker for the formation of the powerful oxidant peroxynitrite, which is presumed to be largely responsible for many of the adverse effects of excessive NO generation (Beckman and Koppenol, 1996; Kristof et al, 1998; van der Vliet et al, 1999; Wizemann et al, 1994).

In addition to a marked increase of iNOS immunostaining intensity in the airways, we observed de novo expression of iNOS in endothelial cells and strong LPS-induced up-regulation of iNOS in VSMC within 1 to 2 hours after LPS application. Such iNOS response in the vascular system appeared thus again much earlier than was shown in cell culture studies, which reported induction of iNOS mRNA within 8 to 24 hours in LPS-treated VSMC (Johnson et al, 1994; MacNaul and Hutchinson, 1993; Nakayama et al, 1992), accompanied by significant peroxynitrite production in this cell type (Kristof et al, 1998). Excessive production of NO in the pulmonary vasculature via iNOS is thought to be a key mechanism in the pronounced systemic vasodilation of sepsis (Gaston et al, 1994, Moncada et al, 1991; Ullrich et al, 1999) and has also been implicated in sepsis-associated pulmonary edema via cytotoxic peroxynitrite formation (Hinder et al, 1999; Kristof et al, 1998). Studies in iNOS-deficient mice showed that these mice were protected from LPS-induced impairment of hypoxic pulmonary vasoconstriction and arterial oxygenation (Ullrich et al, 1999; Weimann et al, 1999) and acute lung injury (Kristof et al, 1998). Thus, inappropriate NO formation from iNOS apparently causes loss of ventilation-perfusion matching including shunt flow in the lung vasculature, and the presently described up-regulation of iNOS in both endothelial and smooth muscle cells well explains such an event in septic lungs.

Summarizing the main immunohistochemical findings, the study shows a constitutive expression of all three NOS isoenzymes in lung tissue with a differential cellular pattern. Differences of NOS isoenzyme expression were particularly noted in endothelial cells, with a baseline expression of eNOS and bNOS and an additional induction of iNOS in response to LPS stimulation. Hilar vein cardiac muscle cells displayed positive staining for bNOS and iNOS, whereas smooth muscle cells of intrapulmonary vessels were positive for all three of the isoenzymes. A different expression pattern was noted in bronchial smooth muscle cells that expressed iNOS and eNOS but not bNOS. Under physiologic conditions bNOS was the only NOS type detected in mast cells; however, in response to LPS, eNOS was found to be de novo expressed. In perivascularly localized leukocytes, iNOS was induced and up-regulated after LPS stimulation. The differential cellular expression and regulation pattern observed may be a result of different cell-specific NOS isoenzyme-dependent cell functions involved in either physiologic or pathophysiologic mechanisms.

In contrast to a previous study on cellular Cox-2 expression in LPS-treated rat lungs (Ermert et al, 2000b), no difference in the cell type–specific expression of eNOS, bNOS, or iNOS between buffer- and buffer/plasma-perfused rat lungs was observed. This is a surprising finding, because CD14-negative cell types, such as bronchial smooth muscle cells, endothelial cells, epithelial cells of small bronchi, and large artery VSMC, showed LPS-induced Cox-2 up-regulation only in the presence of plasma constituents, with LBP and sCD14 assumed to be particularly important (Ermert et al, 2000b). LPS-induced NOS regulation may thus be suggested to engage different signal transduction pathways as compared with Cox-2 regulation in many cell types. Such a difference may include direct gene-regulatory response to endotoxin versus indirect response via LPS-induced cytokine synthesis, eg, TNF-α and IL-1 (Lamas et al, 1991; Nakayama et al, 1992; Robbins et al, 1994), which may then secondarily affect regulation of target systems independent of plasma constituents.

CONCLUSION

In conclusion, the present study demonstrates that LPS challenge of intact lungs exerts marked impact on the NOS regulation in many cell types. On the one hand, strong up-regulation of iNOS is noted in various cells of the vascular and bronchial system, which may provoke inflammatory events via peroxynitrite formation and which may affect the regulation of vascular tone, resulting in loss of ventilation-perfusion matching. On the other hand, eNOS is down-regulated in many cell types, suggesting hindrance of physiologic regulatory events with involvement of eNOS-based NO synthesis. Moreover, differential regulatory profiles as exemplified by eNOS and iNOS expression are noted in many immune-type cells. In contrast, no alteration of bNOS expression was noted in the LPS-treated lungs. Overall, the fundamental changes in cell-specific NOS regulation in response to endotoxin may well contribute to abnormalities in physiologic function in septic lung failure.

Materials and Methods

Reagents

Polyclonal antibodies against bNOS (SA-202), iNOS (SA-200), and eNOS (SA-201) were obtained from Biomol (Hamburg, Germany). Additional antibodies against iNOS and eNOS were purchased from Santa Cruz Biotechnology, Inc. (sc-651, sc-654, both polyclonal antibodies; Heidelberg, Germany) and BD Transduction Laboratories (catalog no. N30020, catalog no. N32020, both monoclonal antibodies; Franklin Lakes, New Jersey). The secondary antibodies goat anti-rabbit F(ab)2 alkaline phosphatase (AP) conjugate and rabbit anti-goat F(ab)2AP conjugate were obtained from Biotrend (Cologne, Germany). The Vector Red Substrate Kit was acquired from Vector Laboratories (Burlingame, California). All other biochemicals were obtained from Merck (Darmstadt, Germany).

Western blots confirming the specificity of the antibodies used in this study were performed with commercially obtained cell lysates and recombinant enzymes recommended for the detection of the specific NOS isoenzymes. For the detection of bNOS, an immunoblotting standard consisting of recombinant brain NOS protein from rat (catalog no. SW-109; Biomol GmbH, Hamburg, Germany) was used. The specificity for eNOS was tested on a recombinant eNOS enzyme isolated from a baculovirus overexpression system in SF9 cells (catalog no. 60880; Cayman Chemical, Ann Arbor, Michigan). For iNOS Western blotting, a cell lysate from RAW264.7 cells (mouse Abelson leukemia virus-transformed macrophages) stimulated with IFN-γ was used (catalog no. sc-2259; Santa Cruz Biotechnology).

Blotting of all three NOS isoforms was additionally performed with a commercial rat brain extract (catalog no. sc-2392; Santa Cruz Biotechnology) and homogenized rat lung tissue. Using the rat brain extract, all three isoforms could be strongly detected with specific bands, as stated in the literature. In the lung tissue homogenate, we detected specific bands with strong expression for eNOS and iNOS, and weak expression for bNOS, probably because the amount of bNOS was lower in the homogenate compared with the other NOS isoenzymes.

Animals

CD rats (Sprague Dawley) were obtained from Charles River (Sulzfeld, Germany). All experimental procedures were performed in conformity with the guidelines of the National Institutes of Health (Guide for the Care and Use of Laboratory Animals, NIH publication no. 86–23, Revised 1985, United States Government Printing Office, Washington DC).

Lung Isolation and Perfusion

The rats (male, body weight 350 to 400 g) were deeply anesthetized with sodium pentobarbital (100 mg/kg body weight ip). After local anesthesia with 2% Xylocaine and a median incision, the trachea was dissected, and a tracheal cannula was immediately inserted. Subsequently, mechanical ventilation was started with 4% CO, 17% O, and 79% N (tidal volume 4 ml, frequency 65/minute, end expiratory pressure 3 cm H2O) using a small animal respirator KTR-4 (Hugo Sachs Elektronik, March, Germany). A median laparotomy was performed, and subsequently the rats were anticoagulated with 1000 U of heparin. After midsternal thoracotomy the right ventricle was incised, a cannula was fixed in the pulmonary artery, and the apex of the heart was cut off to allow pulmonary venous outflow. Simultaneously, pulsatile perfusion with buffer solution was started. The buffer contained 2.4 mm CaCl2, 1.3 mm MgCl2, 4.3 mm KCl, 1.1 mm KH2PO4, 125.0 mm NaCl, and 25 mm NaHCO3 as well as 13.32 mm glucose (pH ranged between 7.35 and 7.40).

The lungs were carefully excised, avoiding any damage, while being perfused with buffer solution, and suspended in an upright position. Next, a cannula was inserted through the left ventricle and fixed in the left atrium to obtain a closed perfusion circuit. Subsequently, lungs were placed in a temperature-equilibrated housing chamber (37° C) freely suspended from a force transducer.

After extensive rinsing of the vascular bed, the lungs were perfused in a recirculating manner with a pulsatile flow of 13 ml/minute. The alternate use of two separate perfusion circuits, each containing 100 ml, allowed the repetitive exchange of perfusion fluid. Perfusion pressure, ventilation pressure, and weight of the isolated organ were registered continuously. The left atrial pressure was set at 2 mm Hg under baseline conditions (0 referenced at the hilum) to guarantee zone III conditions at end expiration throughout the lung.

Lungs selected for the study were those that (a) had a homogeneous white appearance without signs of hemostasis or edema formation, (b) had pulmonary artery and ventilation pressures in the normal range, and (c) were isogravimetric during a steady-state period of 30 minutes.

Experimental Protocol

Rat lungs were perfused for about 5 minutes for washout of blood (n = 5). A total of 60 experiments with isolated rat lung perfusion were performed. In control experiments, rat lungs were perfused for 2 hours solely with buffer fluid (n = 5) or with buffer/plasma (1.5% rat plasma admixed). In additional experiments, LPS was admixed in concentrations of 50 mg/ml, 1,000 ng/ml, or 10,000 ng/ml to the recirculating buffer fluid. Subsequently, both 1-hour and 2-hour perfusion periods were performed. Challenge with LPS was performed in the absence or presence of 1.5% plasma (each group n = 5).

After termination of perfusion, the rat lungs were instilled with TissueTek OCT compound and frozen in liquid N2. The rat lungs were dissected into tissue blocks from all lobes and stored at −80° C. Sections 10 μm in thickness were cut from frozen tissue blocks of both left and right lungs.

Immunohistochemistry

The sections were fixed for 5 minutes with 3% paraformaldehyde solution and washed in PBS (0.01 m, 150 mm NaCl, pH 7.6) for 3 × 5 minutes. They were treated for 15 minutes with a 1% Triton solution. The sections were preincubated in PBS containing 5% goat serum, 1% BSA, and 0.05% Tween-20 to block nonspecific binding. Overnight incubation with the polyclonal primary antibody in PBS containing 1% BSA, 0.05%Tween-20, was performed at 4° C (dilutions: bNOS [Biomol] 1:1000; iNOS [Biomol] 1:3000; iNOS [Santa Cruz] 1:100; eNOS [Biomol] 1:500; eNOS [Santa Cruz] 1:100]. The sections were then washed in PBS and incubated with goat anti-rabbit F(ab)2AP conjugate diluted 1:2000 in the same dilution buffer overnight at 4° C. Afterward, washing of the sections in PBS for 3 × 5 minutes was performed. Subsequently, the sections were developed with a Vector Red Substrate Kit. Levamisole (2.5 mm) was added to inhibit endogenous AP activity. Counterstaining of the sections was performed with hematoxylin. Control staining was performed by omission of the primary antibody and substitution with nonspecific serum at the same dilution.

Image Analysis

The method has been previously described (Ermert et al, 2000a, 2000b, 2001). Briefly, an image analysis system consisting of a 12-bit cooled CCD camera (Sensys KAF 1400; Photometrics, Tucson, Arizona) mounted on a fully automated Leica DM RXA (Leica, Wetzlar, Germany) was used to digitize gray scale images to a Dual-Pentium 200 MHz host computer. Microscope settings were kept constant throughout all measurements (Objective: 40 × oil, Leica PL Fluotar 40 × /0.75). A stabilized 12V tungsten-halogen lamp (100 W) was used for illumination. Microdensitometry was performed with a custom-designed filter for absorbance measurement of the substrate Vector Red (central wave length 525 nm, half-band width 10 nm ± 2 nm) manufactured by Omega Optical, Inc. (Brattleboro, Vermont). Adjustment of all microscope settings was stored and recalled before measurement. Calibration of the measurement system with a reference slide was done before measurement. Gray scale images were digitized to 12-bit accuracy, resulting in an intensity scale ranging from 0 to 4095. Image analysis was performed by means of the image analysis program ImagePro 3.0 (Mediacybernetics, Silver Spring, Maryland). For direct visualization of staining intensity, a pseudocolor scale with 11 colors was chosen, each representing an equal sector of the intensity scale, and applied to the images. Background measurement was performed to evaluate the influence of nonspecific staining and/or absorption of unstained tissue.

Five complete cross-sections per group from both right or left lung were measured. From each section, five images per stained structure (Table 1a, 1b, 2a, 2b, 3a and 3b) were digitized, and the area of interest was manually defined. The mean gray values were automatically measured and subsequently transferred into the spreadsheet program EXCEL (Microsoft Corporation, Redmond, Washington).

Anatomic segments of the vascular tree were defined according to the classification described by Hislop and Reid (1978). In brief, elastic arteries and muscular and partially muscular vessels were distinguished by the structure of the vessel wall. Large arteries at the lung hilum were defined by the thickness of their muscle layer and the occurrence of elastic fibers, whereas hilar veins were identified by the cardiac muscle cells, which accompany the pulmonary veins of rats into the lung parenchyma (Hislop and Reid, 1978). A vessel was termed muscular or partially muscular when smooth muscle cells were observed in the subendothelial layer forming a muscular layer that surrounded the vessel either fully or partly, respectively. The vessels located in the preacinar regions were predominantly of a fully muscular type (associated with bronchioli and terminal bronchioli), whereas within the acinar region (associated with respiratory bronchioli and alveolar ducts) mainly partially muscular and nonmuscular vessels were found in accordance with previous data (Hislop and Reid, 1978).

Statistical Analysis

ANOVA was used to evaluate differences among different groups. A value of p<0.05 was considered significant. Data are given as mean ± sem.