Pulmonary intravascular macrophages (PIMs) are one of the recently recognized representatives of the mononuclear phagocyte system (MPS), a lineage of cells derived from the monoblastic series in the bone marrow (Winkler, 1989; Warner and Brain, 1986; Atwal and Saldanha, 1985). They are most abundantly seen in the lung parenchyma, by virtue of their habitat in the pulmonary microvessels of several species of animals (Atwal et al., 1989; Winkler, 1988). They are recognized by the specific surface-coat and adhesive junctions they form with the pulmonary endothelium, particularly in goat, sheep, cattle, and horse. The surface-coat consists of linearly arranged globular units along the external leaf of the cell membrane, in close conformity to the outer contours of the cell surface. Several experiments have provided indirect evidence that the coat-globules are predominantly composed of lipoproteins (Atwal et al., 1989; Singh et al., 1995a, b; Atwal et al., 1994; Jassal, 1989; Atwal and Minhas, 1992). The linearly-arranged coat-globules are not membrane-bound, but are separated by a translucent gap (the lamina lucida), approximately 35–39 nm wide, from the outer leaflet of the cell membrane.
In general, PIMs satisfy the major functional criterion of marked phagocytosis (Atwal et al., 1992b, c). The phagocytosis of exogenous tracer substance is critically dependent on the integrity of the globular surface-coat (Atwal and Minhas, 1992). Once the surface-coat is washed away by physiological detergents, there is a drastic reduction in the pulmonary clearance of exogenous material injected via the venous channels (Longworth et al., 1996). The PIMs also localize circulating pathogens and concentrate the inflammatory response as putative “hot spots” on the vascular side of the air–blood barrier (Warner, 1996).
In previous studies we have shown that the surface-coat of PIMs in sheep, goat, and cattle is sensitive to a single intravenous heparin treatment as well as to the digestive effect of lipolytic lipase (LPL) in vitro. In concert with heparin-induced translocation of the coat, the plasma levels of oleic, palmitic, and stearic acid were also elevated. After its initial disappearance, the coat is reconstituted 4–8 hr after a single treatment of heparin (Atwal et al., 1989; Atwal et al., 1992a; Singh et al., 1995a). The ultrastructural disappearance of the coat from the surface of the PIMs, 30 min after i.v. injection of heparin, was attributed to the lytic effect of LPL on the globules of the coat. Hormones, cytokines, and heparin are known to release LPL not only from the extracellular matrix of the endothelial cells, but also from monocyte-derived macrophages in tissue culture. Lipolytic lipase (LPL) is also the rate-limiting enzyme in the hydrolysis of serum triglycerides, derived from low-density lipoproteins (LDLs) (Nielsen et al., 1997; Knuston, 2000; Querfeld et al., 1990).
In the present study, we examined the disappearance of the surface-coat of the PIMs in the goat lung after treatment with heparin. Evidence is presented that, in parallel with the disappearance of the coat, there was some enhanced translocation of the globular units of the coat into the vacuolar system of the PIMs. In addition, the Golgi apparatus was seen expanded 2 hr after the heparin administration. Thereafter, the induction of peroxisomes and allied cell organelles took place 4–8 hr after heparin treatment. The peroxisomes reacted positively for catalase after aldehyde fixation and 3,3′-diaminobenzidine (DAB) staining. The DAB technique for localization of catalase, a marker of peroxisomes, is still the cytochemical method of choice in the initial morphological survey for the identification of this organelle (Fahimi and Baumgart, 2000). We also analyzed the formation of peroxisomal reticulum (PR), a transitional organelle thought to be involved in the early stages of de novo formation of peroxisomes (Yamamoto and Fahimi, 1987; Lazarow and Fujiki, 1985).
MATERIALS AND METHODS
Clinically healthy goats (five male and eight female, 6 months to 1 year old (average 9 months)) were used in this study. Goats were purchased from local breeding farms. All animals were acclimatized to the controlled, isolated housing conditions for 1–2 weeks; animals were dewormed, stall fed, and allowed free access to water.
Heparin sodium injection USP was purchased from Allen and Hamburys Co. (Toronto, Canada).
Heparin was injected intravenously at a dose rate of 50 U/g/kg of body weight in the jugular vein of 11 goats. Periods of 30 min (two animals), 2 hr (three animals), 4 hr (three animals), and 8 hr (three animals) were allowed between the injection and killing of the animals. Two animals were injected with normal saline and were used as controls. All animals were euthanized with an overdose of barbiturate.
Immediately after euthanasia the animals were positioned in sternal recumbancy and the lungs were fixed by intratracheal instillation of 500–800 ml of fixative (2% glutaraldehyde and 2% paraformaldehyde in 0.2 M HCl-Na cacodylate buffer, pH 7.4) to an inflation pressure of 30 cm H2O to ensure adequate unfolding of the alveolar wall. The animals were put into a large cooler where fixation was carried out for 30 min for routine electron microscopy and cytochemistry for catalase. After fixation in situ, blocks of tissue were collected from the cranial, middle, and caudal lobes of the right lung. Small strips of tissue were further fixed in the same fixative for 1 hr. Following fixation, the lung tissue was rinsed in the same buffer and cut on a tissue sectioner into 30-μm sections, which were immediately incubated.
Incubation for Catalase
The incubation medium, with minor modifications, was prepared according to the method described by Yamamoto and Fahimi (1987). The medium contained 1 mg/ml of 3,3′-diaminobenzidine tetrachloride (Sigma Chemical Company, St. Louis, MO) prepared in cacodylate buffer at pH 7.4 and 0.02% H2O2. The incubation was carried out at room temperature for 1 hr. For the cytochemical control preparation, we omitted the peroxide substrate in the incubation medium, or boiled the sections for 5 min in glutaraldehyde before incubation (Deimann et al., 1984).
Fixation and Tissue Preparation for Electron Microscopy
After fixation in situ, specimens for routine electron microscopy collected from the cranial and caudal lobes of the left lung and the cranial, middle, and caudal lobes of the right lung were diced into small pieces of about 1 mm3, and fixation was continued by immersion in the same fixative for 1–2 hr.
All tissues taken from all lobes were post-fixed for 90 min in 1.5% OsO4 in 0.1 M HCl-Na-cacodylate buffer (pH 7.6). Tissues for routine electron microscopy were stained en bloc with 0.5% tannic acid in 0.1 M-HCl-Na-cacodylate buffer for 30 min at room temperature.
All tissues thus prepared were dehydrated in ethanol and propylene oxide, and were finally embedded in Jembed 812 (J.B. EM Services Inc.). Thick sections were stained with toluidine blue–basic fuchsin and were viewed in a Zeiss universal microscope. Ultrathin sections for routine electron microscopy were stained with uranyl acetate and lead citrate, whereas ultrathin sections of tissue for cytochemistry (for catalase) were not exposed to double differentiation. All sections thus prepared were examined with a JEOL-100S electron microscope at 80 kV.
We have presented ultrastructural and cytochemical evidence that a single intravenous injection of heparin removed the globular surface-coat and potentiated its partial entry into the vacuolar system of PIMs. Following intracellular translocation of the coat, probably via receptor-mediated endocytosis, the TGN compartment of the Golgi complex expanded as an obligatory response to the intracellular entry of the coat material. In temporal succession to these changes, proliferation of peroxisomes and a dramatic appearance of PR in the PIMs was also seen concurrently with the expansion of ER, which primarily remained closely related to the newly arrived peroxisomes (Yamamoto and Fahimi, 1987; Erdmann et al., 1997). The validity of peroxisomes was confirmed by using the DAB technique for catalase, which is the marker enzyme of peroxisomes. According to Fahimi and Baumgart (2000), the DAB technique for catalase localization is still the most commonly used cytochemical method for the initial morphologic survey to identify this organelle.
In a previous study (Atwal et al., 1992a) we showed that intravenous heparin triggered higher levels of oleic, palmitic, and stearic acids in the plasma. This biochemical change was attributed to the lytic effect of LPL, mobilized from the extracellular matrix of the pulmonary endothelium and/or secreted by the PIMs (Querfeld et al., 1990; Hamosh and Hamosh, 1985; McCrohon et al., 1999). The LPL has an alkaline pH optimum, is activated by a low concentration of heparin, and provides the only way to channel FFAs to the tissues whenever they are needed. The enzyme hydrolyzes the Sn-1 or Sn-3 and Sn-1 ester bonds of triglycerides and phospholipids of lipoproteins, respectively (Hamosh and Hamosh, 1985). The intracellular translocation of the coat-globules can best be explained by assuming that LPL/LDL complexes were internalized by the PIMs upon binding to plasma membrane-associated heparin sulfate, an interaction which is heparin-sensitive under physiological conditions (Schönherr et al., 2000).
Some of the earliest ultrastructural changes 2 hr after intravenous heparin were the expansion of the Golgi apparatus and the proliferation of polyribosomes, monoribosomes, ER, and mitochondria, in combination with the microtubular network. The FFAs play an important role in cell homeostasis by serving as a metabolic energy source and building blocks for membrane lipids, thus regulating various cell functions (Stewart, 2000). Increasing evidence suggests that plasma membrane-associated and cytoplasmic fatty acid binding proteins are involved in the uptake and intracellular transport system of FFAs (Dutta-Roy, 2000; Stewart, 2000; Kennedy, 2000).
The present initial cytoarchitectural changes perhaps underscore the requirement of an upregulated synthesis of binding and cellular transport proteins for the uptake of elevated plasma FFAs by the PIMs (Luby-Phelps and Weisiger, 1996; Weisiger, 1996; Pelham, 1997). These diverse manifestations of nutritional, metabolic, and functional roles of FFAs in the mammalian system depend upon their role in specific targeting of cell organelles, such as the mitochondria, endoplasmic reticulum (McArthur et al., 1999; Ribarik et al., 1999), and Golgi apparatus (Gonatas et al., 1992; Rothman and Orci, 1992; Atwal et al., 1994). The Golgi apparatus serves as a direct pathway for diffuse exchange of lipids. The intermixing (lateral diffusion) of lipid molecules is 10 times faster than that of proteins to reach a maximum intracellular concentration at the Golgi level (VanMeer, 1989; Montgomery and Cohn, 1989; Orci et al., 1997). The conspicuous prominence of the Golgi apparatus (especially the TGN), possibly to an influx of FFAs, 2 hr post-heparin treatment was perhaps an early response, as a relatively resistant cell component for synthesis, metabolism, and membrane trafficking—activities for which the Golgi apparatus appears to be quite labile (Rothman and Orci, 1992; Bannykh et al., 1998; Simons and Ikonen, 2000). It is thought that the attachment of nuclear membrane to microtubular tracts in the cytosol accounts for the nuclear volume changes which occur during cell spreading. Furthermore, mechanical expansion of the nucleus as a result of spreading is thought to open nuclear pores (Luby-Phelps and Weisiger, 1996) necessary for augmented communication with the cytoplasm for anchorage-dependent cell function, particularly cytoskeleton-mediated transport between the nuclear envelope and the Golgi apparatus (Farquhar and Hauri, 1997; Tanaka and Noguchi, 2000).
To our knowledge, the present study demonstrates the first morphologic detection of peroxisomes in the PIMs, which characteristically possess a surface-coat predominantly composed of LDL in goat, sheep, cattle, and horse (Atwal et al., 1994; Atwal and Minhas, 1992; Singh et al., 1995a). The peroxisome is a recognized cell organelle of lipid metabolism (Aboushadi et al., 1992; van den Bosch et al., 1992). In the present model of lipid mobilization, after manipulation with heparin, we also present the induction of a structure reminiscent of PR, a transient precursor structure involved in the biogenesis of peroxisomes (Lazarow and Fujiki, 1985). The PR and peroxisomes were observed in juxtaposition as well as in direct connection with each other, and were exclusively stained for catalase.
In a recent study of HepG2 and Cos-7 cells, images of the dynamic behavior of a variety of morphologically different peroxisomal structures were observed (Schrader et al., 2000). The real-time imaging revealed that moving peroxisomes made transient contacts to form PR. According to these authors, peroxisomal behavior in vivo is significantly more dynamic and interactive than was previously thought (Subramani, 1998). We have observed heterogenous populations of peroxisomal structures and their mutual interactive contacts, although the present interpretation is based on examination of static electron microscopic views of fixed lung tissue. The fusion and fission of microperoxisomes suggest that peroxisome proliferation in the PIMs did take place, and that PR is a transient structure as a relevant source of peroxisome biogenesis (Schrader et al., 2000; Titorenko and Rachubinski, 1998).
There are a number of xenobiotic compounds, hypolipidimic drugs, and eicosanoids that induce peroxisome proliferation (Reddy and Mannaerts, 1994; Reddy and Chu, 1996; Lemberger et al., 1996; Krey et al., 1997). In recent years, among the many inducers of peroxisomal proliferation, FFAs are also recognized as natural inducers. The FFAs, as major dietary constituents, are capable of regulating gene expression in response to food intake and qualitative nutritional changes. The induction of peroxisomes is dependent upon the transcription of responsive genes through the binding of peroxisome proliferator activated receptors (PPARs) (Gearing et al., 1994), which have emerged as one of the central regulators of nutrient/gene interactions.
The early Golgi expansion at 2 hr post i.v. heparin was seen as a cellular requirement of membrane trafficking during cell spreading. It may have acted as a stable compartment of protein (glycoprotein) synthesis for the uptake of FFAs in response to FFA influx and intracellular traffic. Subsequently, the induction of peroxisomes and allied structures was perhaps a response to sustain β-oxidation of increased concentration of FFAs and their metabolic fate, thus modulating cell growth and development via PPAR-γ, as a second stage of this biphasic response (Dutta-Roy, 2000; Krey et al., 1997).
In view of the dynamic relationship of the surface-coat with the membrane of the PIMs, the sensitivity of the surface-coat to heparin (expressed by its rapid mobilization into the cell (Atwal et al., 1992c)), and the elevated plasma levels of FFAs (Atwal et al., 1992a), the induction of peroxisomes could be an FFA-triggered, gene-related encoding of peroxisomal proteins (Gearing et al., 1994). In a recent study, Ricote and coworkers (1999) suggested that FFAs and eicosanoid metabolites activate PPAR-γ, and are involved in macrophage biology and cell cycle regulation. It has already been shown that metabolism of vasoactive lipids in the PIMs is manyfold higher than that of the alveolar macrophages (Bertram et al., 1988).
In addition to their role as phagocytes in the defense against lung disease (Warner, 1996), the present fine structural and cytochemical tests for peroxisomes are suggestive of their capacity to adapt to the circulating level of FFAs in the pulmonary circulation in the goat. The peroxisomal changes in the PIMs are interesting not only in terms of regulating inflammatory disease but also in influencing diet-related lipid homeostasis in the lung. They also raise the possibility of involvement in free-radical and charge-related lung injury, in which activated PIMs may exert an important role (Chang and Voelkel, 1989; Simon et al., 1991).
Finally, the present inquiry is a preliminary one, limited in both extent and depth because of the small number of goats used at each time point. However, we attempted to increase the number of samples from the cranial, middle, and caudal lobes of the right lung at several selective sites for ultrastructural and cytochemical analyses. The changes occurred uniformly and were repeatable at each TEM scrutiny. It is evident that the present model of peroxisomal changes imparts to the PIMs an important role in influencing lipid metabolism in the lung of domestic ungulates. In future research we need to vigorously investigate heparin-induced sequential changes by examining serial sections of lung tissue in order to establish a convincing relationship between PR and peroxisomes. More sophisticated cytochemical techniques for a wider range of peroxisome-specific enzymes in combination with morphometry are available to contribute significantly to our knowledge of peroxisomes in the PIMs (Fahimi and Baumgart, 2000; Usuda et al., 1999). The ubiquity of peroxisomes was firmly established only after the systematic application of the DAB histochemical technique (Fahimi and Baumgart, 2000). The present study is just a beginning in the attempt to define the biology of heparin-induced peroxisomes. So far, in vitro studies have been handicapped because of the technical difficulties involved in the isolation and successful cell culture of PIMs. Until such time as PIMs are successfully isolated with high purity and viability for sensitive microphysiological and biochemical investigations, the present methods of fixation are still the best tools at hand for obtaining cell-specific information related to these unique cells in the lung.