Decreased sialidase activity in alveolar macrophages of guinea pigs exposed to coal mine dust.

The origin of immune dysfunctions that are observed in pneumoconiotic miners still remains unknown. There is evidence that the carbohydrate moiety of membrane glycoconjugates is of primary importance in many functions of immunocompetent cells. The glycosylation, and especially the sialylation level of membrane components of various lymphocyte and macrophage subsets, vary depending on the state of cellular differentiation and activation. Sialidases, which may regulate the amount of sialic acids exposed on the cell membrane, can thus be considered as immunoregulatory enzymes. In this report, the sialidase activity has been measured in alveolar macrophages (AM) and in cell-free bronchoalveolar lavage fluid (BALF) from guinea pigs exposed for 4 months to coal mine dust at a concentration of 300 mg/m3. The samples were collected by bronchoalveolar lavage 2 months after cessation of exposure. The sialidase activity in the cell-free fluid and in the purified alveolar macrophages showed a 10-fold decrease (p less than 0.001). Kinetic parameters of the enzyme such as Km and optimum pH did not change. This changed activity was specific for sialidase, as two other lysosomal glycosidases, beta-galactosidase and N-acetylglucosaminidase, showed unchanged activities. These results suggest the possibility that, by inducing a decreased sialidase activity, exposure to coal mine dust may lead to a modified expression of AM membrane-associated sialic acids giving rise to altered immune functions (i. e., phagocytosis, antigen processing, response to cytokines, etc.).


Introduction
Pneumoconiosis, a chronic fibrotic reaction in the hing, is caused by the inhalation of dust particies contaüüng silica (7,2).In experimental modeis of chronic exposure to mineral dust, it has been shown that alveolar macrophages (AM) play a critical role in the development and persistence ofpulmonary inflammation (3).This inflammation is thought to be partly due to lysosomal enzymes released from lysed AM.Particulate silica ingested by AM has the potential to damage the integrity ofcell membranes, and studies have suggested that lysosomal enzymes are discharged into the cytoplasm after phagolysosomal membrane disruption (4)(5)(6).The mechanisms by which particies are recognized and ingested by AM remain unclear.Evidence suggests that particies bind to cells through electrostatic forces (7).In that context, due to their exposed position on the cell membrane and their negative Charge, sialic acid (SA) residues have been considered important.For instance, sialic acid has the capacity to mediate the initial binding ofpositively charged particies in vitro (8); also, it is implicated in the interacrions ofcoal fly ash with mammalian cells (9).Furthermore, sialic acid residures are important in modulating the biochemical properties ofmany membrane glycoconjugates (10), expressed by immunocompetent cells (II).The sialylation of membrane sialoglycoconjugates is regulated by two Idnds of enzymes, namely, sialyltransferases and sialidases.Factors that modify or alter sialidase activity can influence a diversity of physiological phenomena within cells, on cell surfaces, and in circulating body fluids (72).Sialidases (E.C.3.2.1.18;acylneuraminyl hydrolases) catalyze the hydrolysis of sialic acid residues a-ketosidically bound to oligosaccharides, glycoproteins, and glycolipids.These enzymes are widely distributed in eukaryotic and prokaryotic cells (13}.A large body of evidence suggests that, in eukaryote, the total sialidase activity is actually a composite ofat least three types of activities differing from each other both in their subsü-ate specificity and in their localization within the cell.After tissue homogenization and differential centrifugation, sialidase activities have been demonstrated in the lysosomal, cytosolic, and membrane subcellular fractions; the lysosome-associated form being the predominant one (13).
In a rabbit model, lysosomal enzymes such äs ß-glucuronidase or acid phosphatases are released in the bronchoalveolar lavage fluid (BALF) afier intratracheal mjection ofquartz (14).Because AM possess lysosomal sialidase activity (75) and because this enzyme may participate in the control of the degree of sialylation ofcell surface glycoconjugates resulting in changed biological properties ofthe cells (16,17), we investigated the effect ofcoal mine dust exposure on the sialidase activity in AM and in BALF ofguineapigs.

Bronchoalveolar Lavage
Guinea pigs were sacrificed by IP mjection ofa lethal dose of sodium pentobarbital.The ehest was opened and the blood was collected by cardiac puncture.The trachea was intubated by tracheotomy, and bronchoalveolar lavage (BAL) was then performed with a total lavage volume of 100 mL warmphosphatebuffered saline Dulbecco buffer pH 7.2 (without Ca or Mg 2  ^.Total cell count was performed with a Malassez hemocytometer, and differential cell counts were established on smears stained with May-Grunwald Giemsa (at least 200 cells were counted.) The BAL was centrifüged at 300g for 10 min at 4°C.The BALF supernatant was concentrated 50-fold by filtration on a Diaflo PM 10 membrane (Amicon) at 4°C and then frozen and stored in 0.4-mL aliquots at -80°C.The cell pellet was resuspended in Hanks'saline solution ata concentration of 0.5 X lO'cells/mL, and 20 mL of the Suspension was transferred into 110-mm diameter plastic petri dishes.After a 2-hr incubation at 37°C in a 5 % CO^ humid atmosphere, nonadherent cells were removed and the petri dishes were washed twice with Hanks' saline solution.Adherent cells were harvested with a Teflon scraper, resuspended at a concentration of 10 7 cells/mL in 3 mM imidazol buffer, pH 7.2, containing 250 mM sucrose and l mM ethylenediamine tetraacetate (EDTA).The cells were pelleted by centrifugation at 10,000g for 5 min at 4°C and immediately frozen at -80°C.

Enzymatic Assays
Alveolar macrophages were homogenized using a glass Teflon homogenizer with 20 strokes and assays were carried out immediately.The sialidase activity was assayed by using a fluorogenic Substrate: sodium 4-methylumbelliferyl-Q:-D-neuraminate (Mu-Neu Ac) (79): 20 /xL of cell homogenate or BALF was mixed with 2.5 p,L of2.5 X 10' 3 M Mu-Neu Ac and 2.5 JUL of l M acetic acid/sodium acetate buffer ranging in pH from 4 to 6 and incubated for 2 hr at 37°C.The reaction was stopped with 2.1 mL of 0. l M NaOH.Fluorescence was measured on a Kontron SFM 25 fluorometer (exdtation at 365 nm, emission at 450 nm).Pure methylumbelliferone was used äs a Standard.All assays were done in duplicate.One enzymatic unit was defmed äs the amount of enzyme that hydrolyzes l nmole of Mu-Neu Ac per hourat37°C.
^v-acetylglucosaminidase and jS-galactosidase activities were assayed at their respective Optimum pH using appropriate Substrate (20).Protein concentration was determined using Bradford's method (21) with bovine serum albumin äs a Standard.

Statistical Evaluation
All the results are expressed äs the mean ± SD.Statistical significance was determined by the Student's f-test.

Results
The number of bronchoalveolar cells obtained by lavage from the dust-exposed animals was significantly higher man from the controls(Fig.1),40.07± D.6 and 13.10 ±6.3 X 10" cells, respectively.Differential cell counts showed that the increased total cell number in dust-exposed animals was not attributable to a single cell type.In normal control guinea pigs, 77.6 ± 5.7% (Fig. 2) of the lavaged cells were alveolar macrophages, and 16.2 ± 5.0% were eosinophils; polymorphonuclear neutrophils (PMN) and lymphocytes each represented less man 3 %.In dust-exposed  animals, the relative proportions of PMN increased (5.6 ± 3.6%), whereas those ofthe other cell types were unchanged.On smears stained with May-Grunwald Giemsa, more multinucleated giant cells were observed in dust-exposed animals (2%) than incontrols (< 0.5%).Six months after the beginning of exposure, the protein concentration in BALF was significantly (p< 0.005) increased: 1.65 ± 0.32 and 2.71 ± 1.03 mg/mL BALF from control and exposed animals, respectively.

Sialidase Activity in Guinea Pig Alveolar Macrophages
Sialidase activity, with an Optimum pH at 4.6, was detected in the crude homogenate of control guinea pig alveolar macro- phages (Fig. 3).The mean specific activity was 21.60 ± B. 49 U/mg protein.The apparent Km value for Mu-Neu Ac at pH 4.6 was 1.08 X 10' 4 ± 0.17 x lO^M.The effectof EDTA and of several divalent cations (Zn 3 ranging from 0.5 to 10 x ,Ca , Cu ) at concentrations M using their respective Chlorides (Fig. 4).Cu^ and Zn"" caused a marked inhibition of the sialidase activity.At a 10 mM concentration, Cu^ was able to inhibit about 60 % ofthe initial activity.On the contrary, at low concentration (0.5 mM), Ca 24 ^ and EDTA slightly activated the sialidase activity.

Sialidase Activity in Alveolar Macrophages from Dust-Exposed Animals
To determine ifany quantitative and/or qualitative changes in sialidase content occurred after coal mine dust exposure, the sialidase activity was assayed in AM from 13 dust-exposed guinea pigs.The results obtained are summarized in Figure 5.Control and exposed animals formed two distinct groups, the specific AM sialidase activity being significantly (p< 0.025) decreased in all dust-exposed animals.This drop in sialidase activity reachedupto 90% ofthe initial activity: 2.35 ± 1.47U/mg protein and 21.6 ± 13.49 U/mg protein in the exposed and control group, respectively.Although the sialidase activity decreased in the dust-exposed group, kinetic parameters such äs K( 0.51 ± 0.18 x 10' 4 M) or the Optimum pH at 4.6 ofthe enzyme did not change significantly.
The major sialidase activity is known to be associated with lysosomes.One can assess that the decreased sialidase activity was specific for this enzyme because the two other lysosomal glycosidase activities that were assayed, namely, jS-galactosidase and A^-acetylglucosammidase, were not significantly modified.

Sialidase Activity in BALF
In BALF from normal animals, a sialidase activity with an Optimum pH at 4.6 was detected.A BALF sialidase activity with the same Optimum pH was detected in 12 out ofthe 13 exposed animals.The major change in dust-exposed guinea pigs was a decrease in the total BALF sialidase activity (Fig. 6

Discussion
As assessed by a 3-fold increase in the number of cells harvested by bronchoalveolar lavage, the initial response ofthe lung in guinea pigs that had been exposed to coal mine dust was characterized by an acute inflammatory reaction.As a single intratracheal injection ofpure silica ledto the same observation in guinea pigs (22), silica in coal mine dust is likely to be responsible for this reaction.The local cellular recruitment could be due to the secretion of chemoattractants by dust-loaded AM (23,24).The relative proportion ofAM in the bronchoalveolar cell population of dust-exposed animals was similar to that seen in controls.As compared to controls, more PMN were present in BAL from dust-exposed animals.In addition, the fraction of PMN was slightly but significantly increased.PMN seem to play a critical role in early cellular events occurring in several types ofexperimental modeis ofpneumoconiosis (25,26).Acontroversy still exists conceming the precise role ofPMN during pulmonary diseases.Depending on the experimental model, both protective and causal (27) roles have been attributed to PMN in the development of pulmonary interstitial diseases.
The cellular inflammation was associated with a rise in the BALF protein concentration.In another experimental model where hamsters were exposed to pure silica, a 7-fold increase in the BALF protein concentration was demonstrated (28).Proteins appearing in the airways may originale from various sources including serum exudation, cell lysis (29), local protein synthesis (14), and enzyme secretion.For example, increased lysosomal enzyme activities were found in BAL from mice injected intratracheally with silica (25) and in BAL from silica-exposed workers (30).
In AM-crude homogenate from control guinea pigs, the sialidase activity was Optimum at pH 4.6.Similar observations were reported for rabbit AM (15,31).The limited number of cells recovered by bronchoalveolar lavage did not allow us to study the subcellular distribution of this activity.Previous reports have provided evidence that the bulk of sialidase activity displays acidic Optimum pH and is associated with the fraction enriched in lysosomes (15,(32)(33)(34).A cytosolic sialidase with an Optimum activity at pH 5.4 (75) and a sialidase associated with the plasma membrane, active at pH 4.5 (35), have also been described.As already shown (36), we observed that divalent cations such äs Ca 24 ^ on the one hand, and Cu^ and Zn 24 ^ on the other hand, have opposite effects on the sialidase activity.
Sialidase activity is present in normal guinea pig BALF (37), whereas in humans this enzyme is only present in significant amounts in BALF from patients with interstitial lung diseases such äs idiopathic pulmonary fibrosis or sarcoidosis (38).Nevertheless, to date, the cellular origin ofthis enzyme has not been definitively demonstrated.However, because AM are numerous in bronchoalveolar spaces and because sialidase activities ofAM and BALF have the same Optimum pH, one can speculate that BALF sialidase represents an activity released from AM.
In dust-exposed animals, sialidase activities decreased both in AM and in BALF; however, kinetic parameters such äs K^ and Optimum pH were not modified.Because particulate silica has been shown to alter the cellular metabolism, the decreased BALF sialidase activity may result from a decreased synthesis and/or liberation of cellular sialidase into the respiratory secretions.In the absence ofcell lysis, the decreased BALF sialidase activity can also be the consequence ofan enzymatic inhibition or inactivation.Numerous mechanisms including inactivationby proteases (39) and inhibition by cations (36) may result in a decreased catalytic function of sialidase in BALF from dustexposed animals.
Macrophage sialidase activity can be up-or downregulated.Treatment ofrat peritoneal macrophages with inactivated Streptococcus pyogenes fragments resulted in a 9-fold increase in sialidase activity (35).In vivo bacille Calmette-Guerin (BCG) Stimulation of rabbits induced an upregulation ofAM lysosomal sialidase (75), and a decreased sialidase activity was found in AM from guinea pigs treated with a glycoproteic extract from Klebsiella pneumoniae (37).So far, the precise mechanism of sialidase regulation and its direct effectiveness on the degree of sialylation of sialoglycoconjugates is not fully understood.Modifications of sialidase content with a concurrent decline in membrane-bound sialic acid have been reported.For instance, T-cell activation is accompanied by an increase in the endogenous level of sialidase activity and an hyposialylation of class I major histocompatibility complex antigen molecules (16,17).Previous desialylation of the target cells improves recognition and ingestion by macrophages (40).These results point out the importance of membrane sialic acids in cellular differentiation, activation, and biological properties including phagocytosis.It is therefore conceivable that a sialidase-mediated differential expression of sialic acids on the AM cell membrane may be a mechanism through which impaired cellular fünctions may participate in the development ofpneumoconiosis.A study ofthe relative sialylation ofAM membrane sialoglycoconjugates from either normal control or dust-exposed guinea pigs should help to answer this question.

FIGURE 2 .
FIGURE l.Total cell count in bronchoalveolar lavage fluid from control(n = 6) and dust-exposed (n = 13) guinea pigs.The results are expressed äs means ± SD. (*) Difference between dust-exposed and control animals is significant, p < 0.05; (**) difference between dust-exposed and control animals is highiy significant, p < 0.005.