-
Specificity of Opsonic Antibodies to Enhance Phagocytosis
of Pseudomonas aeruginosa by Human Alveolar Macrophages
HERBERTY. REYNoLDS, JOHNA. KAzMnIEowsyI, and HAoR D H.
NEWBALL
From the Laboratory of Clinical Investigation, National
Institute of Allergyand Infectious Diseases, and Pulmonary Branch,
National Heart and LungInstitute, National Institutes of Health,
Bethesda, Maryland 20014
A B S T R A C T These studies compared the ability ofspecific
secretory IgA (sIgA) and IgG antibodies topromote phagocytosis of
viable Pseudomonas aeruginosaby human alveolar macrophages.
Macrophages wereobtained by lung lavage of normal adult smoker
andnonsmoker volunteers and were maintained as in vitrocell
monolayers. Both immune sIgA and IgG agglu-tinating antibodies were
demonstrated to coat and op-sonize viable bacteria, whereas similar
nonimmune im-munoglobulin preparations did not. When alveolar
mac-rophages were challenged with viable opsonized '4C-la-beled
Pseudomonas, IgG-reacted bacteria were ingestedbetter and killed
more readily than sIgA-opsonizedorganisms. Phagocytic responses
were not significantlydifferent between macrophages obtained from
smokersand nonsmokers. Although sIgA and IgG antibodies canbe found
in respiratory secretions and both are un-doubtedly important in
pulmonary host defense, IgGopsonic antibody was superior in
enhancing the uptakeof Pseudomonas by in vitro-cultured alveolar
macro-phages. It may be the more important respiratoryantibody for
certain bacterial infections.
INTRODUCTIONA complex host defense system protects the lower
res-piratory tract from foreign particles inhaled with ambi-ent
air. Fortunately, various anatomic barriers in theupper airway and
intricate branching of the tracheo-bronchial tree mechanically
exclude particles largerthan 3 Em in diameter from the respiratory
bronchiolesand alveoli (1, 2). However, smaller particles
(0.5-3Mm), which may include infectious agents such as bac-teria,
can be deposited directly in the alveoli (1). Thisinitiates a
complicated, but still poorly understood, in-teraction between
these potentially infectious particles,
Received for publication 19 December 1974 and in revisedform 19
March 1975.
the protein-lipid-enzyme components of the alveolarsurface, and
the resident alveolar macrophages. Ideally,phagocytosis of the
foreign particles by macrophagesbegins the clearance process.
In the present studies, two questions about the con-frontation
between bacteria and macrophages have beenasked: (a) which opsonic
antibody best facilitates bac-terial phagocytosis, and (b) does the
environmentalbackground of the macrophage (i.e. whether the
cellswere obtained from cigarette smokers or nonsmokers)affect
phagocytosis or killing of the bacteria? Wehavechosen Pseudomonas
aeruginosa as the microorganismfor study because it is a frequent
cause of nosocomialpneumonia in patients with cardiorespiratory
diseases(3-5) or with altered immunity (6).
METHODS
Normal subjects. Normal adult volunteers, hospitalizedto
participate in clinical projects at the National Institutesof
Health, underwent limited bronchoscopy for selectivelavage of the
lingula lobe as described previously (7). Bothcigarette smokers and
nonsmokers were used. Informedconsent was obtained. Each subject
was premedicated withintramuscular atropine (0.5 mg) and diazepam
(10 mg);local anesthesia of the respiratory tract was obtained
withtopical 2% lidocaine. Bronchoscopies were performed
trans-nasally with a fiberoptic bronchoscope (model BF-T
5B2,Olympus Corporation of America, New Hyde Park, N. Y.).The
bronchoscope was positioned in the lingula lobe orificeand aliquots
of 50 ml of sterile saline (0.9%o sodium chloride,Abbott
Laboratories, North Chicago, Ill.) were infusedthrough the
bronchoscope and aspirated into a sterile con-tainer. Three 50-ml
saline washings were done.
All subjects were examined regularly during the 48-hinterval
after bronchoscopy. Inspiratory rales were usuallydetected for 2-6
h in the lung area ravaged. All volunteerstolerated bronchoscopy
and lavage well and had no notice-able effects from the
procedures.
Recovery of respiratory cells. The lavage fluid was im-mediately
strained through several layers of very loosecotton gauge to remove
mucus and then centrifuged at 500 g
The Journal of Clinical Investigation Volume 56 August
1975.376-385376
-
for 5 min at 250C. After the supernatant lavage fluid
wasdecanted, the cell pellet was resuspended in modified
Hanks'balanced salt solution (HBSS) 1 (prepared without Ca"+and
Mg++ ions or phenol red), washed, and pelleted twiceby
centrifugation, before a final cell suspension was made.
Cell viability was assessed by eosin Y dye exclusion (8).Cells
were counted (Coulter model F., Coulter ElectronicsInc., Hialeah,
Fla.) and 500 cell differential counts weredone on wet mounts and
cytocentrifuged Wright-Giemsapreparations. To size the various
respiratory cells in theunstained wet mount, the diameter of a cell
was measuredin two planes under oil immersion (1,000 X) and the
aver-age diameter was expressed in micrometers. Macrophageswere
identified by size and morphology, by staining withneutral red (9),
and by ingestion of polystyrene latex balls(Dow Chemical Co.,
Midland, Mich.; mean diameter 1.1gm) .
Establishment of short-term macrophage cell cultures.Respiratory
cells were cultured on glass surfaces in tissueculture chambers
(No. 4802 two-chamber units, Lab-TekProducts, Div. Miles
Laboratories Inc., Naperville, Ill.)with McCoy's 5A medium (Grand
Island Biological Co.,Grand Island, N. Y.) supplemented with 0.3%
vol/volL-glutamine, 10% heat-inactivated fetal calf serum, and
anti-biotics-gentamicin sulfate (5 /hg/ml) and penicillin G(100
U/ml). About 1.5 X 106 viable macrophages were addedto each chamber
and allowed to adhere in an atmosphereof moist air and 5% C02 at
370C. Monolayers were allowedto acclimate for 24 h; then monolayers
were washed threetimes with HBSS, with added Ca`4 and Mg++, and
recon-stituted with 1.9 ml of HBSS/chamber. Adherent cell
mono-layers consisted of approximately 95% alveolar macro-phages.
To insure comparability of cell cultures, individualchambers were
checked for uniformity of the cell mono-layers by examination with
an inverted microscope at 200 X;selected chambers were assayed for
protein by the Lowry,Rosebrough, Farr, and Randall method (10).
Preparation of respiratory secretory IgA antibody. Toobtain
secretory IgA antibody (sIgA), respiratory secre-tions were
collected from a patient, A. W. 0. (NIH#0967865) after natural
Pseudomonas pulmonary infection.The patient had chronic asthma and
two previous episodesof Pseudomonas aeruginosa pneumonia. His
respiratorysecretions were watery and nonpurulent (produced
50-100ml/day), and did not contain detectable IgM immuno-globulin.
High titers of agglutinative antibody activityagainst Pseudomonas
aeruginosa, immunotypes 3 and 7,were present.2 Expectorated
secretions were collected bypostural drainage and immediately
frozen to -20'C with-out preservatives. A pool of 500 ml of
secretions collectedduring a 7-day interval was used in these
experiments. Themethods used to purify sIgA from bronchial
secretions havebeen detailed previously (7). In brief, respiratory
secretionswere emulsified and centrifuged to obtain clear
supernatantfluid (200 ml), which was then dialyzed extensively
againstborate saline buffer, pH 8.0 (12). After further dialysis
in0.02 M Tris-HCl (2-aminohydroxymethyl-1,3-propanediol),pH 8.0,
the secretions were chromatographed on an anion-exchange column of
DE-52 diethylaminoethyl cellulose(Whatman pre-swollen microgranular
DEAE, H. Reeve
'Abbreviations used in this paper: HBBS, Hanks' bal-anced salt
solution; s, secretory.
'Hemagglutination titers were kindly done by Dr. H. B.Devlin,
Research and Development Division, Parke, Davis& Co., Detroit,
Mich., as described (11).
Angel & Co., Inc., Clifton, N. J.) and eluted with a
con-tinuous salt gradient (12). Immunoglobulin-rich materialeluted
at conductivity between 3 and 17 mmhoand pH 8.1-8.7.
Immunoglobulin-containing effluent was concentratedand then
gel-filtered through Sephadex G-200 (PharmaciaFine Chemicals, Inc.,
Piscataway, N. J.) with borate-salinebuffer. In the column
effluent, hemagglutinative antibody ac-tivity was present in the
fractions containing IgA, whicheluted just after the void volume of
the column at about38%o of gel bed volume. These IgA-fractions were
concen-trated and aliquots were purified by ultracentrifugation
in5-ml 10-40% linear sucrose gradients (13). An '25I-labeledrabbit
IgG (14) marker protein was included in each su-crose gradient.
IgA protein had reactivity for secretory piece determi-nants and
a sedimentation coefficient of approximately 11S.Upon
immunoelectrophoresis at a variety of protein con-centrations,
antisera to whole human serum and colostrum(7) detected a single
precipitin arc. An extinction co-efficient of 12.37 (2icm2 nm l%)
was used (15). The finalsIgA preparation of 3.2 mg/ml had a
hemagglutinative titerof 1: 8192 against Pseudomonas immunotype 3
and 1: 2048titer against immunotype 7.
To obtain nonimmune sIgA, respiratory secretions froma patient
(L. C. P., NIH #0995848) with chronic bron-chitis and bronchorrhea,
who had no antibody titers toPseudomonas antigens, were collected
and fractionated asdescribed above. sIgA preparations were stored
at 40C inborate-saline buffer.
IgG antibody isolation from serum. Respiratory secre-tions used
to isolate sIgA antibody contained IgG antibodyas well, which
eluted from Sephadex G-200 at about 52%of gel bed volume. However,
mixed with the IgG weresmall amounts of monomeric 7S IgA (without
secretorycomponent determinants), which could not be
satisfactorilyremoved. Therefore, serum from patient A. W. 0. was
usedas the source of immune IgG opsonin (7); similarly, serumfrom
L. C. P. was used for control nonimmune IgG.
Serum from clotted whole blood was precipitated with30% dry
ammonium sulfate. The serum precipitate was re-dissolved in borate
saline buffer and dialyzed extensively toremove residual ammonium
sulfate before dialysis against0.02 M Tris-HCl. The ammonium
sulfate fraction waschromatographed on DEAEand eluted with a 0.4 M
NaCl-Tris gradient. Low molarity protein, conductivity between1.0
and 9.0 mmho, at pH 8.5 was pooled and concentrated.Further
purification was done in sucrose density gradientsby
ultracentrifugation. An extinction coefficient of 14.3(Z, cm2om,
l%) was used for IgG (16). A final preparationof immune IgG (6
mg/ml) had Pseudomonas aeruginosalipopolysaccharide
hemagglutinative titers of 1:256 againstimmunotype 3 and 7. The
nonimmune IgG at a similar con-centration had titers of < 1:
4.
Antisera and immunologic methods. Rabbit antiseraagainst human
serum, colostrum, and immunoglobulin G andA were obtained from
Behring Diagnostics, AmericanHoechst Corp., Somerville, N. J. Sheep
antiserum to IgAwas supplied by Meloy Laboratories Inc.,
Springfield, Va.Secretory piece determinants on sIgA were detected
withan absorbed sheep antiserum prepared against purified
colos-tral IgA (M33A), kindly provided by Dr. D. S. Rowe
(7).Lyophilized fluorescein-conjugated rabbit antisera to humanIgA
and IgG, absorbed to be heavy-chain-specific, wereobtained from
Behring. These antisera were absorbed ex-tensively with heat-killed
Pseudomonas, type 3, before use.
Specific Opsonic Antibody to Increase Macrophage Phagocytosis
377
-
Immunoelectrophoresis and double-diffusion immunopre-cipitation
were done with standard methods, referred to pre-viously (7).
Radioactive bacteria. Pseudomonas aeruginosa, immuno-type 3 in
the Fisher, Devlin, and Gnabasik typing scheme(17), was obtained as
a lyophilized culture from Parke,Davis & Co. (lot 05074).
Pseudomonas aeruginosa organ-isms were inoculated into 25 ml of
trypticase soy brothcontaining 0.05 mCi L-('C]amino acid mixture
(New Eng-land Nuclear, Boston, Mass.) and grown for 18-20 h at37C.
The bacteria were sedimented by centrifugation at3,000 , for 10 min
and resuspended in 40 ml of 0.85% salineby vigorous mixing. Such
washings were repeated four orfive times until radioactivity in the
supernatant fluid wasminimal. After the final wash, about 90%o of
the radioactivitywas associated with the bacteria. The bacteria
were sus-pended to 10' organisms/ml by optical density at 620
nm;concentrations were confirmed by quantitative pour
platecultures. Bacteria remained in the lag phase of
growththroughout each experiment.
Opsonization of bacteria. Washed suspensions of 'C-labeled
Pseudomonas organisms in saline (108/ml) wereadded to equal volumes
of specific immunoglobulin prepara-tions used as sources of opsonic
antibody. The mixtureswere incubated at 370.C for 15 min. Bacterial
opsonizationwas always done in a slight excess of antibody protein,
andincubation times were limited so that visible aggregation
ofbacteria would not occur. The opsonized bacteria were notwashed
to remove nonreactive protein because centrifugationoften promoted
macroscopic agglutination. Instead, the op-sonized bacteria (0.1-ml
inoculum) were added to an ap-proximately 20-fold excess volume' of
HBISS in the cellculture to dilute unreactive protein to
insignificant amounts.
To insure that bacterial opsonization or coating had oc-curred
with immune IgG and sIgA, three methods wereused to detect antibody
associated with the Pseudomonasorganisms. Viable bacteria (100/0.5
ml) were reacted for15 min with the particular immunoglobulin
preparation,either immune or nonimmune, then pelleted by
centrifugation(3,000 g for 15 min), washed, and finally repelleted
and in-activated with 10 jug/ml concentration of gentamicin.
Thefirst method was to disrupt the Pseudomonas pellet by
soni-cation with a 3: 1 volume of glass powder on ice at 80 Wfor 60
s (Sonifer Cell Disruptor, Model W185D, Ultra-sonic Systems, Inc.,
Farmingdale, N. Y.), and then toexamine the pellet for antibody
with specific antiserum bydouble-diffusion immunoprecipitation.
With the secondmethod, Pseudomonas organisms were reacted with an
equalvolume of fluorescein-conjugated rabbit anti-human IgG orIgA
antiserum (diluted 1: 5 in saline) for 30 min at 370Cand then for
18 h at 4°C. After the bacteria were rewashedthree times, they were
examined for immunofluorescence at500 X (Carl Zeiss, Oberkochen,
Wfirttenberg, West Ger-many, large fluorescent microscope, mercury
arc lamp-HB0200, vertical illuminator with excitation filters of
inter-ference blue KP 490 and KP 500, dichoric reflector 500,and
barrier filter LP 520). In the third method, antibodywas absorbed
to the inactivated Pseudomonas organismsand eluted from the
bacteria with a method modified fromthat described by Eddie,
Schulkind, and Robbins (18). Thebacterial pellet was suspended in 1
ml of 0.1 M citratebuffer, pH 2.2, for 90 min at 370C and then
centrifuged,and the supernate was immediately neutralized with 2
MTris, pH 10.5. The supernate was concentrated to about 0.1ml
volume with negative pressure dialysis and then analyzedwith
specific antisera by double-diffusion immunoprecipi-tation.
Bacterial uptake and killing assay (19). Opsonized radio-labeled
Pseudomonas (0.1 ml) were added to the supernate(1.9 ml HBSS) of
the macrophage monolayers and thechambers were reincubated at 370C
with intermittent shak-ing. The ratio of bacteria to cells was set
at 10: 1. Atvarying intervals duplicate chambers were selected,
thesupernates were decanted, and cell layers were washed
re-peatedly three to four times with 2-ml portions of HBSS.Then the
cell layers were lysed with 2 ml of distilled waterfor 15 min and
scraped with a rubber policeman. A sample(0.1 ml) was aspirated for
bacterial culture and the re-mainder of the cell layer homogenates
was transferred toglass counting vials and dried at 850C. The dried
cellhomogenates were digested with 0.5 ml of 0.2 N NaOHfor 2 h at
370C, after which the mixture was neutralizedwith 0.2 ml of 3%
acetic acid, and 10 ml of Aquasol (NewEngland Nuclear) were added.
Occasionally, about 0.5 mlof distilled water was added to the vials
to improve solu-bility of the sample in the scintillation fluid.
Vials werecounted for 14C activity in a liquid scintillation
spectrometer(Packard Tricarb model 3375, Packard Instrument
Co.,Inc., Downers Grove, Ill.), and net counts expressed incounts
per minute. Vials were counted for 10 min so thatthe standard error
was less than 1%; counting efficiencywas approximately 65%.
The cell homogenate sample for bacterial culture wasserially
diluted in 0.01% human albumin (Abbott Labora-tories, Diagnostics
Div., South Pasadena, Calif.) and dis-tilled water and enumerated
by quantitative pour plates withtrypticase soy agar. The number of
colony-forming bacteriawas counted at 24 and 48 h; the definitive
count was ob-tained from the pour plate having approximately
100colonies.
Monolayers selected for' phagocytic indexes were scannedwith
phase contrast microscopy before the cell chamberswere air-dried
and stained with Wright-Giemsa stain. 500macrophages selected from
at least five high-power micro-scopic fields (oil immersion, 1,000
X) were counted for thepresence or absence of intracellular
bacteria. Those withingested bacteria were divided further into
those with one tofour bacteria per cell and those with five or more
per cell,and these were expressed as a percentage of the number
ofmacrophages with intracellular organisms. The identity ofstained
dishes was unknown during counting, and disheswere chosen at
random.
Thus, to summarize the design of the phagocytosis andbacterial
assay: Cell monolayers were inoculated simultane-ously with
opsonized "4C-labeled Pseudomonas, and duringthe next 60-120 min
duplicate dishes from immune and con-trol groups were selected at
intervals and sampled as de-scribed. The efficiency of the
macrophage monolayer tohandle the bacterial challenge was evaluated
by three param-eters: (a) IC counts in the cell layer homogenate;
(b)number of viable bacteria associated with the cell layer;and (c)
visual estimate of phagocytosis by determining aphagocytic index
from stained cell monolayers taken fromeach group at 30 and 60 min
after inoculation No antibioticswere used in the assay.
RESULTSThe respiratory cell recovery is contrasted for thegroups
of smokers and nonsmokers in Table I. Themean age of the subjects
was 21±+1.0 yr (range 19-23yr). The degree of cigarette smoking for
the smokerswas about 1-3 pack-yr; no volunteers had evidence of
378 H. Y. Reynolds, J. A. Kazmierowski, and H. H. Newbal
-
TABLE I
Cell Recovery and Cell Types in Bronchial Lavage Fluid from
Smokerand Nonsmoker Subjects
Numberrespiratory
Groups cells Viability Macrophages Lymphocytes
X1O6 % % %Smokers 39.0±5.3* 94.241.5 89.8±4.2 7.2±1.8
(n = 5) (24.7-54.0)t (74-96) (4-12)Nonsmokers 13.6±2.3 93.042.3
78.443.1 18.042.2
(n = 6) (6.0-21.1) (72-89) (12-25)
*Mean ± SEM.Range observed for each group.
pulmonary disease and all ventilatory function testswere
normal.
The recovery of infused lavage fluid was about 70±-8.0% (range
50-93%) and was the same for bothgroups. About three times as many
respiratory cellswere obtained from smokers as from nonsmokers,
andthere were significant differences in the cellular com-position
of the lavage fluid. The nonsmokers had agreater percentage of
lymphocytes in the cell differ-ential and a correspondingly lower
percentage of mac-rophages than the smokers (7). However, the
sizedistribution of macrophages by average cell diameter ina wet
cell mount did not differ between smokers andnonsmokers, with about
70% of the macrophages ofintermediate size (13-19 um diameter),
about 20% of10-12 #m in diameter and morphologically similar
toperipheral blood monocytes, and about 5% as giantcells (> 20
um diameter) with multiple nuclei. Poly-morphonuclear cells were
found in less than 3% andbasophil-mast cells were observed very
infrequently.
The number of alveolar macrophages obtained froman individual
was only sufficient, usually, to allow onecomplete bacterial uptake
study comparing two opsoninpreparations. This limitation was
particularly true forthe nonsmokers, from whom fewer respiratory
cellswere recovered.
Evidence that immune IgG and sIgA opsonized(coated) Pseudomonas
organisms. Washed, viabletype 3 Pseudomonas aeruginosa were
incubated withvarying concentrations of immune and nonimmune IgGand
sIgA. Control nonimmune immunoglobulin prepa-rations could not be
detected on bacterial surfaces byimmunofluorescence nor eluted from
these cells. Thus,nonimmune immunoglobulins did not stick
nonspecifi-cally to the Pseudomonas organisms and opsonizationdid
not occur. In contrast, purified immune IgG andsIgA antibodies were
adherent to bacteria, as demon-strated by immunofluoresence, and
could be recoveredfrom aggregated Pseudomonas organisms. The
fluores-cent assay was the most sensitive method used. Immune
IgG, reacted with 10' bacteria in a concentration rangeof
0.2-6.0 mg/ml, was detected on bacterial surfaces at0.4 mg/ml
dilution. sIgA antibody at 0.8 mg/ml gavepositive results with a
comparable number of bacteria.For the phagocytic experiments,
however, greater con-centrations of antibody and fewer bacteria
(108) wereused to insure moderate-to-heavy coating of
bacteria,rather than minimal opsonization just described.
Background counts. In a protein-free medium(HBSS), a large
percentage (50-70%) of glass-ad-herent alveolar macrophages
ingested inert polystyreneballs as a consequence of a process
termed nonimmuneendocytosis (20). In contrast, only about 5% of
macro-phages would phagocytose washed and viable nonopso-nized
Pseudomonas organisms in a HBSSmedium thatdid not contain
additional protein. Therefore, the back-ground uptake of viable
bacteria by alveolar macro-phages was small in the absence of
specific antibodiesor opsonins, and the enhancing effect of any
specificopsonin was usually striking. Another factor contribut-ing
to base-line bacterial counts in this macrophageassay system was
the nonspecific sticking of bacteria tothe glass surface. If washed
SC-labeled Pseudomonaswere added to a culture chamber in 1.9 ml of
HBSSand in the absence of alveolar macrophages, a small butlinear
deposition of bacteria on the glass surface oc-curred. 60 min after
chamber inoculation, about 800-1,000 14C cpm could be recovered
after the glass sur-face was washed several times with medium and
scrapedwith a rubber policeman. Bacteria opsonized with im-mune IgG
or sIgA were no more adherent to the glassthan nonopsonized
ones.
Comparison of immune IgG opsonin and control IgG.The opsonic
effect of immune IgG and nonimmune con-trol IgG of promoting
Pseudomonas uptake by macro-phages obtained from nonsmokers and
smokers was com-pared. Three similar experiments for each category
ofnonsmoker and smoker were performed; representativeresults are
depicted in Figs. 1 and 2. The hemagglutina-tive titer of immune
IgG at the concentration used was
Specific Opsonic Antibody to Increase Macrophage Phagocytosis
379
-
(27%)
(42%"l)o (13%) 6
z -4 ~~~~~~~~~~~~54 b~3 ,2
C
103 0.5 Z0 15 30 60 120
MINUTESAFTER INOCULATION
FIGURE 1 1.5 mg/ml immune (A) and 1.5 mg/ml non-immune (0) IgG
opsonins, isolated from serum, are com-pared. The Pseudomonas,
immunotype 3, inocula incubatedwith the opsonin preparations are
shown above the break inthe long scale for viable bacteria. An
inoculum of 2 X 107/0.1 ml of opsonized Pseudomonas was added to
each macro-phage chamber. Viable bacteria (-) and 'C counts(- -)
for total bacteria in the cell homogenates are shownfor duplicate
monolayers sampled at intervals during a 2-hobservation period.
Phagocytic indexes were calculated at 30and 120 min after
inoculation. Macrophages were obtainedfrom a nonsmoker.
1: 32; the titer of the control IgG was < 1: 2. In Fig.1, the
activity of macrophages from a nonsmoking sub-ject is shown. The
format of this particular experimentdiffers from others in that
observations were continuedfor 2 h after bacterial inoculation of
the cell monolayers.The viability of opsonized bacterial inocula,
shown abovethe break in log scale, left ordinate, were
unchangedduring the assay and indicated that the
Pseudomonasorganisms remained in lag growth phase. The uptake
ofbacteria by both groups of cell monolayers was nearlylinear
during the entire 2-h observation, but the moststriking increase
occurred by 60 min for the immuneIgG-treated organisms. "C
bacterial counts were ap-proximately two times greater at 60 min
for the macro-phages exposed to the immune IgG-treated organismsand
a similar relationship was maintained at completionof the
study.
The higher 'C bacterial counts were corroborated byhigher
phagocytic indexes found at 30 min and 120 minfor the immune
IgG-challenged cell monolayers. Itmight be emphasized that the
phagocytic index was
based on the percentage of macrophages that appearedto have
intracellularly located bacteria, as determinedfrom Wright-Giemsa
cell stains. The dynamics of thephagocytic process were better
appreciated in a prelimi-nary study of the cell monolayers by phase
contrastmicroscopy. Under phase, the continuum between bac-terial
attachment to the macrophage cell surface andsubsequent bacterial
internalization could be followed.Not only did the immune
IgG-Pseudomonas-exposedcell monolayers have more intracellular
bacteria, butthey had many more surface-adherent bacteria than
thecontrol IgG macrophages. Thus, many more bacteriawere associated
with the macrophages in the immuneIgG opsonin group than in the
control group. This in-crease was also reflected in the initially
high bacterialcolony counts obtained for the immune IgG
exposedmonolayers at the 15- and 30-min sampling times.Thereafter
by 60 min, the viable bacterial count had de-creased by more than a
factor of 10. Macrophages wereapparently killing Pseudomonas
organisms more rapidlythan they were being accumulated by the cell
mono-layers. To a much lesser degree the same results wereobtained
with the control IgG-Pseudomonas-exposedmonolayers but in no
instance did any of the parametersof bacterial uptake or killing
approach those obtainedin the immune IgG-exposed monolayers. In
most re-spects, the control IgG-treated Pseudomonas
producedbase-line values similar to those of nonspecific
bacterialdeposition (about 1,500 "C counts at 60 min) and
smallphagocytic indexes.
For contrast, the response of a smoker's alveolarmacrophages is
shown in Fig. 2, with the same IgGpreparations at the same
concentrations described inFig. 1. Observations were continued for
60 min afterbacterial inoculation in this experiment. Total
bacterialuptake was appreciably greater by monolayers exposedto the
immune IgG-opsonized Pseudomonas and at 60min the "C counts were
about four times greater thanthe controls. Likewise, higher
phagocytic indexes werefound for the macrophages infected with the
immuneIgG-treated bacteria. Approximately 10% of the macro-phages
that had ingested immune IgG-coated bacteriaby 60 min contained
five or more organisms per macro-phage; whereas, only 1% of the
cells exposed to thenonimmune opsonized bacteria had as many
intracellularbacteria. Phagocytic indexes for these IgG
controlmonolayers of 5 and 13% at 30 and 60 min were ap-proximately
background values. The viable bacterialcounts (left ordinate) in
the immune IgG-challengedmonolayers increased exponentially during
the first 45min of the assay, but at 60 min viable bacteria had
begunto decrease, despite continued accumulation of 'C bac-terial
counts. The pattern of these assays has shownthat 30-45 min are
required before any bacterial killing
380 H. Y. Reynolds, I. A. Kazmierowski, and H. H. Newball
-
is evident; thereafter, the viable bacterial counts maydecrease
significantly, as shown in Fig. 1.
A comparison of the macrophage responses obtainedwith nonsmoker
(Fig. 1) and smoker (Fig. 2) cellsshows many similar results. The
uptake of immune IgGopsonized "C bacteria was comparable for both
groupsof macrophages, particularly during the first 60 min ofuptake
when the regression slopes of the uptake curveswere almost the same
(0.028±0.006 SE vs. 0.023±0.006). Phagocytic indexes were quite
similar Becausenonimmune IgG failed to opsonize the
Pseudomonaseorganisms adequately, bacterial uptake ("C) and
phag-ocytic indexes were not above background values. Al-though the
viable bacterial counts of immune IgG-re-acted Pseudomonas at 60
min were lower for non-smoker macrophages (Fig. 1) than for
smokers' macro-phages (Fig. 2), this apparently enhanced
bacterialkilling was not substantiated in other experiments.
Inessence no striking differences were found to distinguishthe
responses of smoker and nonsmoker alveolar macro-phages.
10'
I-
m 105
0
E
10z
00
SMOKERALVEOLARMACROPHAGESImmune IgG Opsonin (1.5 mg/'ml)
-aControl IgG Opsonin (1.5 mg/ml)
(44%)(22Yo(
(13 Y
0 2 5 151,k1t,-e
30 45 60
MINUTESAFTER INOCULATION
FIGuRE 2 The response of alveolar macrophages from asmoker is
shown with immune and control IgG-reactedtype 3, "C-labeled
Pseudomonas (2 X 107/0.1 ml inoculum/chamber). Duplicate monolayers
from each group weresampled at intervals after inoculation for
viable colony-forming bacteria (left ordinate) and total
["C]bacteria(right ordinate). Phagocytic indexes are given in
paren-
theses.
LAJ0
C;
0z-I0
C)w
2
104
103
102 -6002 5 15 30 45
MINUTESAFTER INOCULATION
3
20.54x
1 aQ0.5 40
FIGURE 3 The ability of 0.8 mg/ml immune (0) and 0.7mg/ml
nonimmune (0) sIgA to promote alveolar macro-phage uptake of
Pseudomonas aeruginosa, immunotype 3,is shown. Viable ["C]bacteria
were added (1.7 X 107/0.1 ml)to the supernatant fluid of cell
monolayers; duplicate cham-bers were assayed at intervals after
inoculation. Solid linesdenote the number of colony-forming
bacteria cultured fromthe cell homogenates; dashed lines show total
["Cibacteriain the cell homogenates as radioactivity on the right
ordi-nate. The percentage of macrophages in the monolayer hav-ing
phagocytized bacteria at 60 min is given in the paren-theses
(phagocytic index). Macrophage viability at the endof the 60-min
observation period was 90%. Cells werelavaged from a smoker.
Studies with sIgA. The effects of immune and non-immune
secretory IgA antibody to enhance bacterialuptake were evaluated
with macrophages obtained fromtwo smokers. One such comparison is
shown in Fig. 3.The IgA preparations were compared at
approximatelythe same protein concentrations. Each monolayer
wasinoculated with a 10: 1 ratio of opsonized Pseudomonasto
macrophages, and duplicate chambers were sampledfrom each group at
2 min after infection and at 15-minintervals for 1 h. Bacteria, as
shown by "C counts,were taken up by the cell monolayers in roughly
linearfashion and about a thousand "C counts were accumu-lated by
both groups of cell monolayers at 60 min. Al-though more viable
Pseudomonas were cultured fromthe immune sIgA-exposed monolayers,
the phagocyticindexes were comparable for each group. In
essence,the immune sIgA only gave a modest increase in macro-phage
uptake, compared with the control. It should beemphasized that
Pseudomonas reacted with sIgA anti-body did have demonstrable
antibody identified andeluted from the bacteria, so that
opsonization in factoccurred. In contrast, nonimmune sIgA did not
coatthe Pseudomonas.
Specific Opsonic Antibody to Increase Macrophage
Phagocytosis
,I II
I,-,
-I- - - --.,,,
1., --
I--,I--,
381
-
103 02 5 15 30 45 60MINUTES AFTER INOCULATION
FIGURE 4 1.4 mg/ml immune sIgA (0), isolated fromrespiratory
secretions, and 1.2 mg/ml immune IgG (A),isolated from serum, each
had hemagglutinative antibodyactivity for Pseudomonas, immunotype
3. The two antibodypreparations were compared at approximately
equal proteinconcentrations for opsonic activity. An inoculum of
1.5 X107/0.1 ml of opsonized viable 'C-labeled Pseudomonas wasadded
to each monolayer. Duplicate monolayers from eachgroup were
selected at intervals during the next 60 minfor culture of viable
bacteria (-) and for total "C bac-terial counts in the cell
homogenates (--- ). Respectivephagocytic indexes are shown in
parentheses. Macrophageswere lavaged from a smoker.
These results with immune and nonimmune sIgA aresimilar to the
background values obtained with this ex-perimental system.
Nonspecific bacterial attachmentto the glass surface of the
macrophage chamber willresult in comparable 'C bacterial counts,
despite re-peated and vigorous washing of the macrophage
mono-layers to remove extraneous bacteria. Phagocytic in-dexes of
about 5-8% can be calculated when washednonopsonized, viable
Pseudomonas are added.
A direct comparison of immune sIgA and immuneIgA opsonins was
made with macrophages obtainedfrom three subjects: two smokers and
a nonsmoker. Inthe representative experiment illustrated by Fig. 4,
al-veolar cells were obtained from a smoker. Both opsoninswere used
at approximately the same protein concen-trations; IgA had an
agglutinative titer of 1: 1,600 andIgG a titer of 1: 32 against
immunotype 3 Pseudomonas
aeruginosa, the same organism used in the other ex-periments.
During the 60 min of this assay, the XC bac-terial uptake was twice
as high in the monolayers ex-posed to IgG-treated bacteria as in
the sIgA-challengedgroup. This difference was reflected in the
higher phago-cytic indexes calculated at 30 and 60 min for the
IgGgroup. Phase contrast microscopy revealed many moremacrophage
cell surface-associated bacteria in the IgG-exposed monolayers.
From the stained monolayer prepa-rations made at 60 min, 14% of the
IgG-exposed macro-phages with intracellular Pseudomonas
(phagocyticindex 40%) had ingested five or more bacteria per
cell.In contrast, only 2% of the IgA challenge
macrophages(phagocytic index 17%) contained five or more bac-teria.
The number of viable bacteria cultured from thecell homogenates was
higher at each sampling intervalfor monolayers exposed to
IgG-opsonized bacteria, con-sistent with the greater number of
total "C bacteriafound associated with this group. However, at 60
minthe colony counts decreased for the IgG-challenged
cellmonolayers, suggesting that Pseudomonas killing wasbeginning;
such a finding was not observed for theIgA-exposed cells.
Therefore, it was clear that sIgAwas less effective than IgG in
promoting bacterial up-take under these experimental
conditions.
Two additional points should be considered. First,sIgA has been
examined in this macrophage assay sys-tem at various protein
concentrations that ranged from0.8 mg/ml (Fig. 3) to the maximum
concentration of3.2 mg/ml. Results with immune sIgA-opsonized
bac-teria were not improved with respect to control nonim-mune sIgA
or to immune IgG values. Second, whensIgA concentrations greater
than 2.0 mg/ml (or ag-glutinative titer 1: 2,048) were used in
opsonizationstudies, viability of the Pseudomonas inoculum (100/0.1
ml) decreased about 50% during the 60-min intervalof the
experiment, because of presumed bacterial clump-ing. If clumps of
bacteria were added to the cell culturesupernate, macrophage
phagocytosis was generally less.
DISCUSSION
An important aspect of this study was the attempt toidentify the
immunoglobulin class that provides themost efficient opsonic
antibody for certain bacterial in-fections of the lung. This
finding could influence theroute of immunization for bacterial
vaccination or theproducts used for passive antibody
administration. Al-though interpretation of the results must be
limitedbecause of the in vitro design of the experiments andthe use
of a single bacterial species, efficacy of IgGantibody to enhance
Pseudomonas uptake was apparent.
The specific immunologic recognition of IgG-coatedviable
bacteria must be distinguished from the non-immunologic
phagocytosis (20) of inanimate particles
382 H. Y. Reynolds, J. A. Kazmierowski, and H. H. Newball
167F
-
such as polystyrene balls. Less than 10% of the alveo-lar
macrophages in a monolayer would ingest washedviable Pseudomonas
organisms in a protein-free me-dium; whereas more than 50% ingested
polystyreneballs under similar conditions. With appropriate
op-sonization with immune IgG, 30-40% of the macro-phages would
ingest viable Pseudomonas within 30-60min after bacterial
challenge. In contrast, nonimmuneor control IgG does not
effectively interact or coat thebacteria, and subsequent
phagocytosis is no better thanthat obtained with unsensitized
Pseudomonas in a pro-tein-free medium (5-13%). Alveolar macrophages
havecell surface receptors for IgG (21, 22) and are able
tophagocytose IgG antibody-coated erythrocytes and bac-teria. Thus,
the results obtained with IgG-opsonizedPseudomonas were largely
predictable from knowledgeof the specificity of the antibody and
the appropriatereceptor of the alveolar macrophages.
sIgA-opsonized Pseudomonas behaved in vitro as ifalveolar
macrophages did not have an attachment siteor specific receptor for
this immunoglobulin. Admittedlythis conclusion is inferred, because
IgA receptors onalveolar macrophages have not been studied, to
ourknowledge. Alternatively, the fault could lie with
theopsonization potential of the sIgA antibody. Recentstudies (23)
have found that i1S human sIgA antibody(anti-A isoagglutinins
purified from colostrum) didnot opsonize type A erythrocytes for
phagocytosis byeither monocytes or polymorphonuclear
leukocytes.Poor sIgA coating of the Pseudomonas organisms wasnot
considered a problem in our studies. After briefantibody reaction
with the Pseudomonas, as done for15 min in these studies, and
washing the bacteria,sIgA could be still detected immunologically
on theorganisms. Thus, at best, after 60 min of opportunity,only
13-17% of monolayer macrophages ingested sIgA-opsonized Pseudomonas
(Figs. 3 and 4). In addition,bacterial-macrophage surface
attachment, as viewed withphase contrast microscopy, was much less
evident thanthat seen with IgG-opsonized bacteria.
Macrophage interaction with IgM-opsonized Pseudo-monas was not
included in these studies because it hasbeen amply documented (7,
24-26) that lower respira-tory tract secretions of normal subjects
do not containdetectable amounts of IgM. Furthermore, the
unlikelyimportance of IgM opsonic antibody in the diseasedlung is
illustrated by the fact that human alveolarmacrophages do not have
a cell membrane receptor forIgM antibody and are, therefore, unable
to ingest IgM-coated particles (21). In addition,
polymorphonuclearleukocytes that readily enter infected pulmonary
tissuedo not have IgM receptors either (23, 27).
In a healthy human, the IgG present in pulmonarysecretions may
originate from two sites, local synthesis
in the submucosa of the respiratory tract or the intra-vascular
globulin pool. With immunofluorescent tech-niques, IgG-producing
plasma cells have been identifiedin human respiratory tissues (28,
29) in numbers ap-proximately equal to IgA cells. Moreover, cell
culturesof bronchial wall and lung tissues are capable of
syn-thesizing IgG and IgA proteins (30). In addition,
in-travascular IgG apparently diffuses into respiratorysecretions.
If this passive diffusion in humans is analo-gous to that observed
in animals, about 1% of a paren-teral dose of homologous 'I-labeled
IgG can be re-covered from rabbit lung washings (31). In
normaldogs, about 0.1% of the parenteral dose of 'I-labeledIgG is
recovered in serial bronchial washings (J. Kaz-mierowski,
unpublished observations). These experi-mental diffusion studies of
IgG only provide some ra-tionale for the well-documented clinical
use of periodicgammaglobulin injections in patients with hypo-
oragammaglobulinemia (32). These passive immuniza-tions seemingly
decrease the incidence of bacterial pul-monary infections in these
patients. Although commer-cial gamma globulin preparations consist
primarily ofIgG globulin (33), IgG antibody may be the most
im-portant component. Its superior opsonic activity andspecific
receptor attachment to lung macrophages couldbe the explanation for
its efficacy in reducing bacterialpulmonary infections.
Without question, alveolar macrophages lavaged fromcigarette
smokers differ from those recovered fromnonsmokers. Many parameters
vary: absolute quantity
(7, 26, 34-36), cytoplasmic inclusions altering morphol-ogy (34,
36-38), quantity of intracellular enzymes (35,38), and higher
resting glucose-energy requirements(34). However, phagocytosis of
Staphylococcus albus(34) and heat-killed Candida albicans or
Aspergillusfumigatus spores (39) has not been different.
Recently,however, Yeager, Zimmet, and Schwartz (36) notedthat
pinocytosis of ["C]sucrose was decreased insmokers' macrophages. In
our studies, we did not finda noticeable difference in the uptake
of IgG-opsonizedPseudomonas by alveolar macrophages obtained
fromsmokers and nonsmokers (compare Figs. 1 and 2). Itis possible
that subtle differences were minimized be-cause of our decision to
culture the macrophages invitro for at least 24 h before bacterial
challenge. Thisroutine was established because it was found that
freshglass-adherent alveolar macrophages (2-3 h after bron-chial
lavage) invariably formed cell surface rosetteswith IgG-opsonized
erythrocytes (EA) and did notphagocytose these immune complexes
(21). It re-quired 24-30 h of in vitro acclimization with our
ex-perimental culture conditions before erythrophagocytosisof EA
complexes became maximal. Harris, Swenson,and Johnson (34) used
freshly lavaged alveolar macro-
Specific Opsonic Antibody to Increase Macrophage Phagocytosis
383
-
phages in phagocytosis studies with Staphylococcusalbus mixed in
autologous serum. They found that onlysmall numbers of macrophages
(18.5-23%) had phago-cytized bacteria at the end of a 3-h
incubation period.In contrast, Cohen and Cline (39) demonstrated
thatwith prolonged in vitro culture for many days therewas a
progressive increase in phagocytic capacity ofalveolar macrophages
that was a function of increasingcell size. To get a high
percentage of macrophages toingest microorganisms for satisfactory
phagocytic ex-periments, a period of cell adjustment to in vitro
con-ditions is necessary. However, in this interval macro-phages
might dispel metabolic products that were per-haps toxic in the
original alveolar environment; thus,initial differences in cellular
function could be over-looked between smokers and other
controls.
ACKNOWLEDGMENTSThe authors appreciate the review of the
manuscript by Drs.Charles H. Kirkpatrick, Anthony S. Fauci, and
Sheldon M.Wolff.
REFERENCES1. Green, G. M. 1970. The J. Burns Amberson
Lecture-
in defense of the lung. Am. Rev. Respir. Dis. 102: 691-703.
2. Green, G. M. 1973. Alveolobronchiolar transport mecha-nisms.
Arch. Intern. Med. 131: 109-114.
3. Tillotson, J. R., and A. M. Lerner. 1968. Characteristicsof
nonbacteremic Pseudomonas pneumonia. Ann. Int.Med. 68: 295-307.
4. Rose, H. D., M. G. Heckman, and J. D. Unger. 1973.Pseudomonas
aeruginosa pneumonia in adults. Am. Rev.Respir. Dis. 107:
416422.
5. Graybill, J. R., L. W. Marshall, P. Charache, C. K.Wallace,
and V. B. Melvin. 1973. Nosocomial pneu-monia: a continuing major
problem. Am. Rev. Respir.Dis. 108: 1130-1140.
6. Pennington, J. E., H. Y. Reynolds, and P. P. Carbone.1973.
Pseudomonas pneumonia. A retroactive study of36 cases. Am. J. Med.
55: 155-160.
7. Reynolds, H. Y., and H. H. Newball. 1974. Analysis ofproteins
and respiratory cells obtained from human lungsby bronchial lavage.
J. Lab. Clin. Med. 84: 559-573.
8. Hanks, J. H., and J. H. Wallace. 1958. Determinationof cell
viability. Proc. Soc. Exp. Biol. Med. 98: 188-192.
9. Cohn, Z. A., and E. Weiner. 1963. The particulate hy-drolases
of macrophages. II. Biochemical and morpho-logical response to
particle ingestion. J. Exp. Med. 118:1009-1019.
10. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R.J.
Randall. 1951. Protein measurement with the Folinphenol reagent. J.
Biol. Chem. 193: 265-275.
11. Crowder, J. G., M. W. Fisher, and A. White.
1972.Type-specific immunity in pseudomonas diseases. J. Lab.Clin.
Med. 79: 47-54.
12. Reynolds, H. Y., and J. S. Johnson. 1971. Structuralunits of
canine serum and secretory immunoglobulin A.Biochemistry. 10:
2821-2827.
13. Reynolds, H. Y., R. E. Thompson, and H. B. Devlin.1974.
Development of cellular and humoral immunity in
the respiratory tract of rabbits to Pseudomonas
lipo-polysaccharide. J. Clin. Invest. 53: 1351-1358.
14. Reynolds, H., M. Carta-Sorcini, and R. Mage. 1973.The
immunoglobulins derived from a VH-CHa recom-binant rabbit and its
normal relatives. I. Measurementsof heavy chain molecular weights
and allotypes of colos-tral IgA. Immunochemistry. 10: 443-447.
15. Newcomb, R. W., D. Normansell, and D. R. Stand-worth. 1968.
A structural study of human exocrine IgAglobulin. J. Immunol. 101:
905-914.
16. Little, J. R., and H. Donahue. 1968. Spectral propertiesof
proteins and small molecules of immunological in-terest, in Methods
Immunol. Immunochem. 2: 343-364.
17. Fisher, M. W., H. B. Devlin, and F. J. Gnabasik. 1969.New
immunotype schema for Pseudomonas aeruginosabased on protective
antigens. J. Bacteriol. 98: 835-836.
18. Eddie, D. S., M. L. Schulkind, and J. B. Robbins. 1971.The
isolation and biologic activities of purified secretoryIgA and IgG
anti-Salmonella typhimurium "O" anti-bodies from rabbit intestinal
fluid and colostrum. J. Im-munol. 106: 181-190.
19. Reynolds, H. Y., and R. E. Thompson. 1973. Pulmonaryhost
defenses. II. Interaction of respiratory antibodieswith Pseudomonas
aeruginosa and alveolar macrophages.J. Immunol. 111: 369-380.
20. Rabinovitch, M., and M. J. De Stefano. 1973.
Particlerecognition by cultivated macrophages. J. Immunol.110:
695-701.
21. Reynolds, H. Y., J. P. Atkinson, H. H. Newball, andM. M.
Frank. 1975. Receptors for immunoglobulin andcomplement on human
alveolar macrophages. J. Im-munol. 114: 1813-1819.
22. Golde, D. W., T. N. Finley, and M. J. Cline. 1974.The
pulmonary macrophage in acute leukemia. N. Engi.J. Med. 290:
875-878.
23. Zipursky, A., E. J. Brown, and J. Bienenstock. 1973.Lack of
opsonization potential of llS human secretoryyA. Proc. Soc. Exp.
Biol. Med. 142: 181-184.
24. Keimowitz, R. I. 1964. Immunoglobulins in normalhuman
tracheobronchial washings. J. Lab. Clin. Med.63: 54-59.
25. Falk, G. A., A. J. Okinaka, and G. W. Siskind.
1972.Immunoglobulins in the bronchial washings of patientswith
chronic obstructive pulmonary disease. Amer. Rev.Respir. Dis. 105:
14-21.
26. Sharp, P. M., G. A. Warr, and R. R. Martin. 1973.Effects of
smoking on bronchial immunoglobulins. Clin.Res. 21: 672.
(Abstr.)
27. Ross, G. D., M. J. Polley, E. M. Rabellino, and H. M.Grey.
1973. Two different complement receptors onhuman lymphocytes. J.
Exp. Med. 138: 798-811.
28. Tourville, D. R., R. H. Adler, J. Bienenstock, and T.B.
Tomasi, Jr. 1969. The human secretory immunoglobu-lin system:
immunohistological localization of 'yA secre-tory "piece" and
lactoferrin in normal human tissues.J. Exp. Med. 129: 411429.
29. Martinez-Tello, F. J., D. G. Braun, and W. A. Blanc.1968.
Immunoglobulin production in bronchial mucosaand bronchial lymph
nodes, particularly in cystic fibro-sis of the pancreas. J.
Immunol. 101: 989-1003.
30. VanFurth, R., and F. Aiuti. 1968. Immunoglobulin syn-thesis
by tissues of the gastro-intestinal and respiratorytracts. Protides
Biol. Fluids, Proc. Colloq. Bruges. 16:479484.
31. Reynolds, H. Y., and R. E. Thompson. 1973. Pulmonaryhost
defenses. I. Analysis of protein and lipids in bron-chial
secretions and antibody responses after vaccination
384 H. Y. Reynolds, J. A. Kazmierowski, and H. H. Newball
-
with Pseudomonas aeruginosa. J. Immunol. 111: 358-368.
32. Fudenberg, H. H., R. A. Good, H. C. Goodman, W.Hitzig, H. G.
Kunkel, I. M. Roitt, F. S. Rosen, D. S.Rowe, M. Seligmann, and J.
R. Soothill. 1971. Primaryimmunodeficiencies: report of a World
Health Or-ganization committee. Pediatrics. 47: 927-946.
33. Heiner, D. C., and L. Evans. 1967. Immunoglobulinsand other
proteins in commercial preparations of gam-maglobulin. J. Pediatr.
70: 820-827.
34. Harris, J. O., E. W. Swenson, and J. E. Johnson, III.1970.
Human alveolar macrophages: comparison ofphagocytic ability,
glucose utilization and ultrastruc-tures in smokers and nonsmokers.
J. Clin. Invest. 49:2086-2096.
35. Cantrell, E. T., G. A. Warr, D. L. Busbee, and R. R.Martin.
1973. Induction of aryl hydrocarbon hydroxylase
in human pulmonary alveolar macrophages by cigarettesmoking. J.
Clin. Invest. 52: 1881-1884.
36. Yeager, H., Jr., S. M. Zimmet, and S. L. Schwartz.1974.
Pinocytosis by human alveolar macrophages: com-parison of smokers
and nonsmokers. J. Clin. Invest. 54:247-251.
37. Pratt, S. A., M. H. Smith, A. J. Ladman, and T. N.Finley.
1971. The ultrastructure of alveolar macro-phages from human
cigarette smokers and nonsmokers.Lab. Invest. 24: 331-338.
38. Martin, R. R. 1973. Altered morphology and increasedacid
hydrolase content of pulmonary macrophages fromcigarette smokers.
Am. Rev. Respir. Dis. 107: 596-601.
39. Cohen, A. B., and M. J. Cline. 1971. The human
alveolarmacrophage: isolation, cultivation in vitro, and studiesof
morphologic and functional characteristics. J. Clin.Invest. 50:
1390-1398.
Specific Opsonic Antibody to Increase Macrophage Phagocytosis
385