Copyright CERS Journals Ltd 1993
European Respiratory Journal
ISSN 0903 • 1936
Eur Resplr J, 1993, 8, 48H88
Printed In UK • all rights reserved
Spontaneous heat shock protein synthesis by alveolar
macrophages in interstitial lung disease associated
with phagocytosis of eosinophils
B.S. Polla*1", S. Kantengwa*, G.J. Gleich**, M. Kondo*, C.M. Reimert*, A.F. Junodt
Spomaneous heat shock protein synthesis by alveolar macrophages in interstitial lung
disease associated with phagocytosis of eosinophils. B.S. Polla, S. Kantengwa, GJ.
G/eich, M. Kondo, C.M. Reimert, A.F. Junod. ©ERS Journals Ltd 1993.
ABSTRACT: Synthesis of heat shock proteins (HSPs) is induced in all cells and
tissues after exposure to elevated temperatures, or a variety of other types of In·
jlll·y, Including oxidat.ive Injury. We have p1•eviously reported that stress proteins
arc induced In monocytes-macrophages during phagocytosis of red blood cells.
Receptor-medlated phagocytosis i.s associated with activation of the respiratory burst,
generation of the lipid mediators of inflammation, and increased production of
cytoklnes. Similar activation events have been described In the alveolar macrophage
(AM) during pulmonary fibrosis.
We therefore analysed the pattern of proteins synthesized by human AMs recovered by bronchoalvcolar lavage (BAL) In lnterstjtiaJ lung disease, both under
ba<ial conditions and after in vilro exposure to heat or hydrogen peroxide (H,01). In
two out of the 17 cases studied, we observed a high alveolar eosinophilia (10 and
24%, respectively) and phagocytosis, by the AMs, of eosinophilic mate.rial. Whereas
exposure to heat or Hp1 Induced In all AMs the synthesis of the classical HSPs,
In these two cases, we found spontaneous synthesis of HSPs and of a 32 kD ox.idation-speclflc stress protein, baeme oxygenase (HO). Exposure of AM to purified
eosinophil-derived proteins, such as major basic prote.ln (MBP), eosinophil peroxi·
dase (EPO), eosinophil-derived neurotoxin (EDN), alone or in conlbination, did not
Induce stress protein synthesis, fw1her suggesting that phagocytosis is Involved In
this Induction.
Stress protein synthesis by AMs may represent a new cellular marker of pulmonary irijury and eosinophilic inflammation, and an autoprotective mechanism against
oxidative stress.
Eur Respir J. 1993, 6, 483-488.
Exposure of cells to environmental stresses, such as
elevated temperatures or reactive oxygen species (ROS),
results in the induction of a specific set of proteins, the
heat shock, or stress, proteins (HSPs) (reviews [1, 2]).
HSPs are not solely induced by stress, but also during
physiological cellular events, such as differentiation, cell
cycle, or cell activation (1]. We have investigated the
relationships between oxidative injury and the heat shock
response [3]. Exogenous hydrogen peroxide (J\OJ in·
duces in human monocytes and alveolar macrophages the
synthesis of HSPs (3, 4], and HSPs are also synthesized
during endogenous generati.on of ROS by the phagocyte
itself. Phagocytosis of sheep erythrocytes by human
monocytes and alveolar macrophages (AMs) is associated
with the synthesis, by the phagocytic cells, of the classical
HSPs, as well as of a 32 kD oxidation-specific stress
protein, haeme oxygenase (HO); both phagocytosis and
generation of ROS are required for this induction (5].
• Allergy Unit, Division of Immunology
and Allergy, and t Respiratory Division,
University Hospital, 1211 Geneva 14,
Switzerland. •• Dept of Immunology,
Mayo Clinic, Rochester, Minnesota 55905,
USA. • Lab. Med . Allergology,
Rigshospitalet University Hospital, Copenhagen N, Denm8lk.
Correspondence: B.S. Polla
Allergy Unit, University Hospital
CH-1211 Geneva 14
Switzerland.
Keywords: Eosinophil, haeme oxygenase,
oxidative injury, reactive oxygen species,
respiratory burst, stress proteins.
Received: July 31 1992
Accepted after revision October 15 1992
This work was supported by grants from
the Swiss National Research Foundation
(3.960.0.87 and 32-028645.90 to BSP), the
Centre de Recherches m~dicales C8llos et
Elsie de Reuter, and NIH (USA) grant If
no. A I 09728. SK was supported by
grants from the Sir Jules Thorn Ch8litsble
Trust and the Helmut Horten Foundation.
Given the involvement of HSPs in thermotolerance and
their role as antigens, HSP synthesis in this context may
either serve as an autoprotective mechanism [6], or participate in inflammation and autoimmunity (7]. Indeed,
antibodies or T-cells specific for HSPs have been found
in a variety of inflammatory and/or autoimmune diseases
(review [8)). HSP-specific T-cells appear to be predominant at the site of ongoing inflammation; for such
T-cells to play a role in pathology, however, one must
assume that host cells at the site of inflammation synthesize HSPs [9).
Interstitial pulmonary fibrosis is characterized by AM
activation, by an increased production of cytokines and
ROS, and by an ongoing immune response (10, 11]. We
investigated the protein synthesis of human AMs recovered by bronchoalveolar lavage (BAL) performed for di·
agnostic purposes in 17 patients with diffuse interstitial
lung disease. Whereas AMs recovered from all patients
B.S. POUA ET AL.
484
synthesized, in vitro, the classical HSPs upon exposure to
heat shock or to H2 0 2 , in two patients with alveolar
eosinophilia, we observed striking spontaneous ex vivo
synthesis of HSPs and HO. We then investigated the
effects of purified eosinophil-derived toxic proteins, in
particular eosinophil peroxidase (EPO) and major basic
protein (MBP), on the stress response of AMs in vitro.
The results suggest that phagocytosis of eosinophils is
required for stress protein induction.
Patients
The clinical characteristics and BAL results of 17
consecutive patients undergoing BAL during the diagnostic work-up of interstitial lung diseases are summarized in table 1. Transbronchial biopsies were obtained
in 10 cases, necropsy was performed in three, and diagnosis relied on clinical presentation and bronchoalveo·
Jar lavage alone in four. One patient was excluded from
the series because of bacterial contamination of the BAL
material. Patients with lung cancer were used as negative controls and for the in vitro experiments with the
purified eosinophil proteins [12]. None of the patients
had eosinophilia on white blood cell differential.
Methods
BAL cells were processed and AMs isolated and pu·
rifled as described previously [12]. Cells were prepared
for microscopy, and stained with May-Griinwald-Giemsa
and nonspecific esterase: 200 cells per slide were
counted, on two separate preparations, and by two independent workers not involved in the present research.
Photographs were taken with a Zeiss Axiophot micro·
scope (x750). Exposure to heat and Hz02 were as described previously [4, 13]; cells were then allowed to
recover for 2 h at 37°C before labelling. For protein
analysis, purified AMs were incubated in RPMI 1640
without methionine (Gibco, Paisley, Scotland), with or
without fetal calf serum (FCS) for 2 h, and then labelled
with 12 J!Ci-mJ·1 L-35S-metbionine (Amersham, Buckinghamshire, UK) for 90 min at 37°C, washed twice with
phosphate buffered saline and lysed in lysis buffer (14].
Proteins from aliquots corresponding to equal cell numbers were then resolved using sodium dodecyl sulphate
polyacrylamide gel electrophocesis (SOS-PAGE) in slab
gels with 10% acrylamide (14]. Quantitative appraisal
of labelled proteins was performed by scanning densitometry of autoradiographs (Genoscan TM laser beam
densitometer, Genofit, Geneva, CH).
Table 1. - Clinical characteristics and BAL results of 17 consecutive patients undergoing BAL during the diagnostic
work-up of interstitial lung diseases
Case
no.
Sex
Age
yrs
1
2
F
F
37
37
3
4
M
F
F
M
M
38
53
43
56
80
M
M
M
M
M
M
M
M
M
F
51
24
28
40
26
32
83
38
63
74
5
6
7
8
9
10
11
12
13
14
15
16
17
Diagnosis
HP
Atypical
pneumonia
Sarcoidosis
IPF
CTD
Cardiopathy•
Pulmonary
embolism••
IPF
Sarcoidosis
Sarcoidosis
Sarcoidosis
Sarcoidosis
Sarcoidosis
HP•••
Pneumonia
••••
IPF
Actinic
f ibrosis
Total BAL
cells xl0 6
Macrophages
%
Lymphocytes
%
Neutrophils Eosinophils HSPs on
%
%
SDS-PAGE
48
76
65
98
27
3
5
2
0
0
40
11.7
65
33
9
5
33
28.8
12
8
13.3
41
25
30
250
62
57
90
29
6
68
12
86.5
52
84
56
85
93
13
43
0.5
5
6
3.5
13
1
3
2
92
20
66
14
4
1.5
12
36
4
4
52
4
94
81.5
45
83
2
1.5
31
0
6.5
0.5
24
10
2.5
8
(+)
+++
+++
(+)
0
0
(+)
(+)
8
3.5
1
0
0
(+)
+
(+)
+
3
2
7.5
HSPs on SDS-PAGE: comparative appreciation of the amount of HSP70 synthesized by AMs under basal conditions (·:
undetectable; (+)/+: presence of a faint but detectable band at 70 kD, as for example in case no. 6, see lane 1, fig. 2; +++:
amount of HSP70 corresponding to the bands observed at 70 kD in cases no. 4 and 5, see lanes 2 in fig. 3a and b). HSP: heat
shock protein; SDS·PAGE: sodium dodecyl sulphate polyacrylamide gel electrophoresis; AM: alveolar macrophage; IPF: idiopathic
pulmonary fibrosis; HP: hypersensitivity pneumonitis; CTD: connective tissue disease; BOOP: bronchiolitis obliterans with organizing pneumonia. •· patient no. 6 died suddenly probably from arrhythmia. He had previously recovered from bronchitis.
IPF was suspected on the basis of the clinical presentation but was not firmly established. • • · patient no. 7 died from heart
failure; necropsy revealed massive pulmonary embolisms and histology confirmed underlying IPF. •• •· patient no. 13 was
chronically exposed to spores from fungi and endotoxins. A suspected diagnosis of histiocytosis X (7% OKT6 positive cells in
bronchoalveolar lavage) was not confirmed on lung biopsy and electron microscopy. ••••· patient no. 15 presented with unilateral
patchy infiltrates, for which no definitive diagnosis has yet been established.
HEAT SHOCK PROTEINS IN ALVEOLAR MACROPHAOES
Exposure to eosinophilic peptides
AMs were isolated as described above and cultured in hydroxylethylpiperazine ethanesulphonic acid
(HEPES)-buffered methionine-free medium without FCS.
Eosinophil peroxidase (EPO) and major basic protein
(MBP) were purified to physical homogeneity from
the eosinophils of patients with the hypereosinophilic
syndrome as described previously [15, 16], and were
used at a final concentration of 10 ~-tg·mJ· 1 , either alone
or combined with 10"" M HP:t and 1Q-1 M NaCl. The
cells were incubated with these compounds for 18 h at
37°C and then labelled with 9 ~-tCi·mJ· 1 35S-methionine for
90 mi n. Proteins were analysed as described above.
Eosinophil-derived neurotoxin (EDN) and eosinophil
cationic protein (ECP) were purified as described and
used at 5 and 10 IA&·ml' 1, respectively [17, 18]. Incubation was for 3 h. For peroxidase staining, AM smears
were successively immersed in benzidine, benzidine~02
solution (1:1) and ·stained with 10% Giemsa solutiOn
(Fluka).
Results and Discussion
Stress proteins include several families of proteins, such
as HSPs and HO. In normal cells, these proteins play
important roles in protein folding, assembly, and translocation, whereas in stressed cells, these functions are
further elicited and are part of the cell's protective and
adaptative mechanisms [1-3). In two patients (cases no.
4 and 5) with interstitial lung inflammation, we found
high alveolar eosinophilia (24 and 10%, respectively), and
phagocytosis of eosinophils by the AMs (fig. 1 A and
B); there was spontaneous synthesis of high levels of
HSPs by these latter cells. In the other cases, including
485
one with 7.5% and two with 8% eosinophils (cases no.
17, 7 and 10), but without phagocytosis of eosinophilic
material/eosinophils by the AMs (not shown), protein
synthesis was similar to that found in cases without BAL
eosinophilia, which suggests that the phagocytic process
is required for HSP induction.
Figure 2 shows protein synthesis by AMs from one
representative case with low alveolar eosinophilia (case
no. 6, 2.5% eosinophils), either under basal conditions
(lanes 1 and 4), or after exposure to 45°C for 20 min
(lanes 2 and 5), or Hp 2 (5x1Q-3 M, 120 min) (lanes 3
and 6). AMs were labelled as described, eilher in the
presence (6a) or absence (6b) of 5% fetal calf serum.
Exposure to heat shock induced the synthesis of the
classical HSPs, 65, 70, 83 and 110 kD (arrows). Exposure to Hp2 resulted in similar qualitative induction
of HSPs, although quantitatively less important, as is
usually observed with oxidative, as compared to thermal,
stress [3] (compare lanes 2 and 3, and 5 and 6). There
was no effect of serum on HSP synthesis after heat
shock. The decrease in overall protein synthesis following exposure to ~02 was, however, more pronounced in
the absence of serum, which is probably due to the lack
of 0 2 radical scavengers normally present in serum.
In several cases, we observed a low level of spontaneous synthesis of HSP70 in AMs from patients without
alve<llar eosinophilia (see for example fig. 2, lanes 1 and
4), probably related to the stress imposed on the cells by
the isolation and culture procedures themselves, as has
been described previously for type Il pneumocytes [19].
This low grade HSP synthesis ((+)/+ in table 1, last lane
(HSPs on SDS-PAGE)) was however not constant, and
not comparable to the high levels synthesized either after exposure to heat (fig. 2, lanes 2 and 5), or spontaneously in phagocytosing AMs (fig. 3A; lanes 2 and 3; fig.
3B, lane 2).
Fig. 1. - Phagocytosis, by the alveolar macrophages (AMs), of eosinophilic material (A. case no. 4), or whole eosinophils (B, case no. 5).
Alveolar cells were prepared for microscopy as described under methods. Anows indicate free alveolar eosinophils and intracytoplasmic eosinophilic
material in AM. (Magnification x750).
B.S. POLIA ET AL.
486
1
2
3
4 5
6
A
1
2
3
-4- 1 10 kD
+---11 0 kD
+--+--+---
70kD
83 kD
-4-
70 kD
-4-
65k0
-4-
32 kD
65 kD
C
B
6a
-4-
83kD
4a
4l>
2
3
6b
Fig. 2. - Protein synthesis by AMs after exposure to heat shock or
H,02• Autoradiographs of SOS-PAGE (10% polyacrylamide gels) of
cell lysates of AMs from case no. 6. The cells were labelled either
in the presen.ce (a) or in the absence (b) of S% fetal calf serum.
Lanes 2 and 5: heat shock; lanes 3 and 6: H,01 ; lanes 1 and 4:
controls. Arrows point to the major HSPs (6S, 70, 83 (83-90) and
110 kO). AM: alveolar macrophage; SOS-PAGE: sodium dodecyl
sulphate polyacrylamide gel electrophoresis; HSP: heat shock protein.
Figure 3 shows the spontaneous synthesis of the classical HSPs (65, 79, 83-90 and 110 kD), and of a 32 kD
protein identified as HO, in comparison with the pattern
of induction by cadmium or erythrophagocytosis [2, 5]
by the AMs from cases no. 4 (fig. 3a) and 5 (fig. 3b),
as compared to a control case (c) (without interstitial lung
disease). With the exception of cases no. 4 and 5, HO
was not detected under basal conditions or after exposure
to heat or exogenous J\Oz alone.
The levels of HSPs synthesized under basal conditions
in these two cases was similar to that observed after ex·
posure to heat in other cases. Spontaneous synthesis of
stress proteins by the AMs may represent a new marker
for injury to these cells during inflammatory lung diseases. This synthesis is probably related to phagocytosis
and the associated generation of high levels of ROS. The
fact that in the AMs HO was observed only under conditions of ongoing phagocytosis is consistent with our
previous observations on the requirements of high intra·
cellular ROS production for this oxidation-specific protein to be induced. We have previously reported that
erythrophagocytosis by AMs induces the synthesis of the
classical HSPs as well as of HO; both phagocytosis and
generation of oxygen radicals appeared to be required for
this induction, since neither activation of the respiratory
burst by non-phagocytic stimuli nor phagocytosis of inert material (latex beads) had such effects (5].
._. 110 kD
. _ . 83 kD
._. 70kD
._. 65kD
. _ . 32kD
c
Sa
5b
Fig. 3. - Protein synthesis by AMs from cases no. 4 and S.
Autoradiographs of SOS-PAGE (10% polyacrylamide gels) of cell
lysates of AMs from case no. 4 (A) and 5 (B). In case no. 4, 4a
and 4b indicate labelling for 90 min (4a) or overnight (4b); in case
5, Sa and 5b correspond to labelling under basal conditions (Sa) or
after heat shock (Sb). Arrows point to the major HSPs and HO (32
leD). C: control; HO: haeme oxygenase. For further abbreviations
see legend to figure 2.
Phagocytosis of eosinophils by AMs may represent a
mechanism for clearing these toxic cells from the alveolar space (20], as has been described for neutrophils [21 ].
We hypothesized that the toxic products of the eosinophil
itself (MBP, ECP, EDN and EPO) could participate in
the stressful events responsible for stress protein induction. Exposure of AMs to the purified eosinophil-derived
487
HEAT SHOCK PROTEINS IN ALVEOLAR MACROPHAGES
proteins ECP, EDN, MBP and EPO did not induce HSP
synthesis, even under conditions in which there was toxicity to the AMs, as assessed by morphological criteria
(AM vacuolization) and a decrease in total normal protein synthesis (not shown). In order to verify the uptake of EPO by AMs, peroxidase staining was performed .
AMs incubated with MBP became vacuolated but remained peroxidase negative, whereas AMs incubated with
the mixture EPO!Hp;NaCI became peroxidase positive,
indicating uptake of EPO by AMs. These findings indicate that EPO alone, or in combination with MBP, is not
sufficient for the induction of a stress response in human
AMs. The spontaneous HSP induction that we observed
in some cases of interstitial diseases is, thus, probably
mediated by the phagocytosis of eosinophils (including all
of their toxic products), rather than by selected eosinophil-derived toxic proteins. The question as to whether
the toxic effects of eosinophils and/or their products in
other diseases associated with pulmonary or bronchial
eosinophilia, such as asthma, also leads to stress protein
synthesis by other (non-phagocytic) injured cells (e.g.,
bronchial epithelial cells) remains to be answered.
The cases presented here do not yet allow us to determine whether HSP synthesis by AMs is, or is not,
strictly associated with phagocytosis of eosinophils by this
cell, or with a particular type of inflammatory lung disease. It will be necessary to investigate broader groups
of patients, with or without eosinophils in BAL, as well
as other forms of eosinophilic pneumonitis. In three patients with Carrington's disease (eosinophilic pneumonitis, without eosinophil phagocytosis), no particuJar HSP
expression was found in bronchoalveolar cells using
specific anti-HSP70 monoclonal antibodies and electron
microscopy (A. Janin er a/; unpubHshed data), which
further stresses rhe apparent specifici ty of HSP synthesis
in pulmonary inflammation associated with phagocytosis
of eosinophils by the A M. In vitro studies testing
whether ingestion of senescent eosinophils induces HSP
synthesis by the phagocytosing AMs will be required to
prove this issue.
HSPs may exert several dictinct functions in inununity.
They may participate in local inflammatory processes,
either by serving as self antigens or by participating in
antigen processing [8). In order, however, for self HSPs
to become targets for the immune system, these intracellular proteins must become available for T-cell recognition. This may indeed be the case during inflammation;
JAJUOUR et al. [22] suggested cell surface expression of
HSP70 on AMs from one patient with interstitial fibrosis. Alternatively, stress proteins may represent an autoprotective mechanism against oxidative injury in the lung.
Indeed, stress proteins have a general function in cellular protection from injury [23, 24} and we have previously reported that HSPs have the potential to protect
monocytes-macrophages from oxidative injury (4, 13].
Furthermore, HO has a recognized antioxidant potential
(25], and heat shock has also been shown to protect cells
from cytolysis by tumour necrosis factor [26], a cytokine
potentially involved in inters titial lung diseases (11].
Finally, it is possible that the association of spontaneous
stress protein synthesis by AMs with alveolar eosinophilia
relates to the observation that an increased number of
eosinophils in BAL defines a group of patients suffering
from pulmonary fibrosis with a poor response to therapy,
with corticoresistance [27], and with progressive lung
disease [28, 29).
Acknowledgements: The authors are grateful to C.
Reynaud and G. Werlen for their participation in the early
phases of this work.
References
1. Morimoto RI, Tissieres A, Georgopoulos C. - The stress
response function of the proteins, and perspectives. /n:
Morimoto RI, Tissieres A, Georgopoulos C, eds. Stress Proteins in Biology and Medecine. New York, Cold Spring
Harbor Laboratory Press, 1990; pp. 1-36.
2. Welch WJ. - The mammalian stress response: cell
physiology and biochemistry of stress proteins. In: Morimoto
RI, Tissieres A, Georgopoulos C, eds. Stress Proteins in Biology and Medecine. New York, Cold Spring Laboratory
Press, 1990; pp. 223-278.
3. Donati YRA, Slosman DO, Polla BS. - Oxidative
injury and the heat shock response. Biochem Pharmacal 1990;
40: 2571-2577.
4. Polla BS, HeaJy AM, Wojno WC, Krane SM. - Hormone let, 25-dihydroxy-vitarnin D3 modulates heat shock response in monocytes. Am J Physiol 1987; 252 (Cell Physiol,
21}: C640-C649.
5. Clerget M, Polla BS. - Erythrophagocytosis induces heat
shock protei.n synthesis by human monocytes-macrophages.
Proc Natl Acad Sci USA 1990; 87: 1081-1085.
6. Kantengwa S, Donati YRA, Clerget M, et al. - Heat
shock proteins: an autoprotective mechanism for inflammatory
cells? Sem lmmunol 1991; 3: 49-56.
7. Polla BS, Kantengwa S. - Heat shock proteins and
inflammation. Curr Top Microbiol/mmuno/1991; 167: 93-105.
8. Young RA, Elliott TJ. - Stress proteins, infection, and
immune surveillance. Cell 1989; 59: 5-8.
9. Koga T, Wand-Wiirttenberger A, DeBruyn J, et al. - Tcells against a bacterial heat shock protein recognize stressed
macrophages. Science 1989; 245: 1112- 1115.
10. Crystal RG, Bitterman PB, Rennard SI, Hance AJ, Keogh
BA. - Interstitial lung diseases of unknown cause: disorders
characterized by chronic inflammation of the lower respiratory
tract. N Engl J Med 1984; 310: 154-166.
11. De Rochemonteix-Galve B, Dayer J-M, Junod AF. Fibroblast-alveolar cell interactions in sarcoidosis and idiopathic
pulmonary fibrosis: evidence for stimulatory and inhibitory
cytokine production by alveolar cells. Eur Respir J 1990; 3:
653-{)64.
12. Dayer JM, Sundstrom L. PoUa BS, Junod AF. - Resolution of factors produced by human alveolar macrophages in culture which stimulate fibroblast proliferation, production of
collagenase or prostaglandin E1 . J Leuk Bio/1985; 37: 641-649.
13. Polla BS, Bonventre JV, Krane SM. - 1,25-dihydroxyvitamin D3 increases the toxicity of hydrogen peroxide in
the human monocytic line U937. J Cell Bioi 1988; 107: 373380.
14. Laemmli UK. - Cleavage of structural proteins during
assembly of the head of bacteriophage T4• Nature 1970; 227:
68()....{585.
15. Gleich GJ, Loegering DA, Mann KG, Maldonado JE. Comparative properties of the Charcot-Leyden crystal protein
and the major basic protein from human eosinophils. J Clin
Invest 1976; 57: 633-{)40.
488
B.S. POUA ET AL.
16. Ten RM, Pease LR, McKean DJ, Bell MP, Gleich GJ. Molecular cloning of the human eosinophil peroxidase. Evidence for the existence of a peroxidase multigene family. J
Exp Med 1989; 169: 1757-1769.
17. Reimert CM, Venge P, Kharazmi A, Bendtzen K. Detection of eosinophil cationic protein (ECP) by an enzymelinked immunosorbent assay. J Jmmunol Methods 1991; 138:
285-290.
18. Reimert CM, Minuva U, Kharazmi A, Bendtzen K. Eosinophil protein X/eosinophil derived neurotoxin (EPX/EDN):
detection by enzyme-linked immunosorbent assay and purification from normal human urine. J Immunol Methods 1991;
141: 97-104.
19. Brandes ME, Finkelstein FN. - Induction of the stress
responses by isolation of rabbit type II pneumocytes. Exp Lung
Res 1989; 15: 93-111.
20. Stern M, Meagher L, Savill J, Haslett C. - Apoptosis in human eosinophils. Programmed cell death in the
eosinophil leads to phagocytosis by macrophages and is
modulated by IL-5. J lmmunol 1992; 148: 3543-3549.
21. Shellito J, Sniezek M, Wamock M. - Acquisition of
peroxidase activity by rat alveolar macrophages during pulmonary inflammation. Am J Pathol 1987; 129: 567-577.
22. Jarjour W, Tsai V, Woods V, et al. - Cell surface expression of heat shock proteins. Arthritis Rheum 1989; 32
(Suppl. 4): S44 (Abstract).
23. Mizzen LA, Welch WJ. - Characterization of the
thermotolerant cell. 1. Effects on protein synthesis activity
and the regulation of heat-shock protein 70 expression. J Cell
Bioi 1988; 106: 1105-1116.
24. Landry J, Lamarche S, Chretien P. - Heat shock proteins: a lead to understanding of cell thermotolerance. In:
Henle KJ, ed. Thermotolerance and thermophily. Boca Ratan,
FL, CRC Press, 1987; 1: pp. 145-173.
25. Keyse SM, Tyrrell RM. - Heme oxygenase is the major 32 kDa stress protein induced in human skin fibroblasts
by UVA radiation, hydrogen peroxide, and sodium arsenite.
Proc Natl Acad Sci USA 1989; 86: 99-103.
26. Jaattelli M, Saksela K, Saksela E. - Heat shock protects WEHI-164 target cells from the cytolysis by tumor
necrosis factors a and~. Eur J Immuno/ 1989; 19: 1413-1417.
27. Polla BS, Perin M, Pizurki L. - Regulation and functions of stress proteins in allergy and inflammation. Clin Exp
Allergy 1992, (in press).
28. Davis WB, ·Fells GA, Sun X-H, et al. - Eosinophilmediated injury to lung parenchyma! cells and interstitial matrix. A possible role for eosinophils in chronic inflammatory
disorders of the lower respiratory tract. J Clin Invest 1984; 74:
269-278.
29. Peterson M, Monick M, Hunninghake GW. - Prognostic role of eosinophils in pulmonary fibrosis. Chest 1987; 92:
51- 56.