Immune Reconstitution during Pneumocystis Lung Infection:
Disruption of Surfactant Component Expression and Function
Elena N. Atochina-Vasserman,2* Andrew J. Gow,†Helen Abramova,* Chang-Jiang Guo,†
Yaniv Tomer,* Angela M. Preston,‡James M. Beck,‡and Michael F. Beers2*
Pneumocystis pneumonia (PCP), the most common opportunistic pulmonary infection associated with HIV infection, is marked by
impaired gas exchange and significant hypoxemia. Immune reconstitution disease (IRD) represents a syndrome of paradoxical
respiratory failure in patients with active or recently treated PCP subjected to immune reconstitution. To model IRD, C57BL/6
mice were selectively depleted of CD4?T cells using mAb GK1.5. Following inoculation with Pneumocystis murina cysts, infection
was allowed to progress for 2 wk, GK1.5 was withdrawn, and mice were followed for another 2 or 4 wk. Flow cytometry of spleen
cells demonstrated recovery of CD4?cells to >65% of nondepleted controls. Lung tissue and bronchoalveolar lavage fluid
harvested from IRD mice were analyzed in tandem with samples from CD4-depleted mice that manifested progressive PCP for
6 wks. Despite significantly decreased pathogen burdens, IRD mice had persistent parenchymal lung inflammation, increased
bronchoalveolar lavage fluid cellularity, markedly impaired surfactant biophysical function, and decreased amounts of surfactant
phospholipid and surfactant protein (SP)-B. Paradoxically, IRD mice also had substantial increases in the lung collectin SP-D,
including significant amounts of an S-nitrosylated form. By native PAGE, formation of S-nitrosylated SP-D in vivo resulted in
disruption of SP-D multimers. Bronchoalveolar lavage fluid from IRD mice selectively enhanced macrophage chemotaxis in vitro,
an effect that was blocked by ascorbate treatment. We conclude that while PCP impairs pulmonary function and produces
abnormalities in surfactant components and biophysics, these responses are exacerbated by IRD. This worsening of pulmonary
inflammation, in response to persistent Pneumocystis Ags, is mediated by recruitment of effector cells modulated by S-nitrosylated
The Journal of Immunology, 2009, 182: 2277–2287.
important determinants of susceptibility to the pathogen. Diseases
affecting cell-mediated immunity, such as leukemia, SCID, and
HIV infection, are classically associated with PCP, but the infec-
tion has been reported as a complication of a wide range of other
immunosuppressive conditions (reviewed in Refs. 4, 5). Respira-
tory failure continues to represent a major sequel of PCP. While
neumocystis pneumonia (PCP)3has long been recognized
as an opportunistic infection affecting immunocompro-
mised patients (1–4). Host defense factors are considered
alveolar infiltrates appear on the chest radiograph, the degree of
hypoxemia, altered lung mechanics, and increased work of breath-
ing are often out of proportion to the severity of the radiographic
findings, suggesting the existence of intrapulmonary shunting and
microatelectasis. Work from this laboratory and others has pro-
vided evidence that PCP, both directly and via the host inflamma-
tory response, induces dysfunction of the pulmonary surfactant
Pulmonary surfactant is a surface-active mixture of phospho-
lipid and protein secreted by alveolar type II cells that reduces
surface tension at the air-liquid interface, permitting alveolar sta-
bility throughout the respiratory cycle (12, 13). In addition to lipid,
biochemical analyses have characterized four unique surfactant-
associated proteins (SP) designated SP-A, SP-B, SP-C, and SP-D
(reviewed in Refs. 14, 15), which can be subdivided into two
groups based on structure, function, and solubility. The hydropho-
bic proteins SP-B and SP-C are involved in adsorption of lipid at
the air-liquid interface (16, 17). The more hydrophilic proteins,
SP-A and SP-D, do not contain significant intrinsic biophysical
activity but are members of a growing family of proteins that plays
a role in the innate or non-Ab-mediated immune response (18).
The term collectin (collagen-like lectin) has been used to describe
this family whose non-lung members include serum proteins man-
nose-binding lectin, bovine conglutinin, and CL-43 (19). SP-A and
SP-D are recognized as important components of the pulmonary
innate immune system modulating complex interactions that occur
between pathogens and host effector cells (reviewed in Ref. 20).
SP-D is a single gene product that can be detected both in type
II cells and Clara cells (21, 22). Monomeric SP-D (43 kDa) con-
tains a collagen-like triple helical domain and a calcium-dependent
*Division of Pulmonary, Allergy, and Critical Care Medicine, Department of Medi-
cine, University of Pennsylvania School of Medicine, Philadelphia, PA 19104;†De-
partment of Pharmacology and Toxicology, Ernest Mario School of Pharmacy, Rut-
gers University, Piscataway, NJ 08854; and‡Division of Pulmonary and Critical Care
Medicine, Department of Internal Medicine, University of Michigan Medical School,
and Veterans Affairs Medical Center, Ann Arbor, MI 48109
Received for publication August 22, 2008. Accepted for publication December 8,
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported by National Institutes of Health Grants R01 HL64520 (to
M.F.B.) and R01 HL083482 (to J.M.B.).
2Address correspondence and reprint requests to Dr. Michael F. Beers and Dr. Elena
N. Atochina-Vasserman, Pulmonary and Critical Care Division, University of Penn-
sylvania School of Medicine, H410F Hill Pavilion, 380 S. University Avenue, Phil-
adelphia, PA 19104. E-mail addresses: firstname.lastname@example.org and atochina@
3Abbreviations used in this paper: PCP, Pneumocystis pneumonia; SNO, S-nitroso-
thiol; BALF, bronchoalveolar lavage fluid; iNOS, inducible nitric oxide synthase;
IRD, immune reconstitution disease; LA, large aggregate; SA, small aggregate; SP-A,
pulmonary surfactant protein A (35 kDa); SP-B, pulmonary surfactant protein B (9
kDa); SP-C, pulmonary surfactant protein C (3.7 kDa); SP-D, pulmonary surfactant
protein D (43 kDa).
The Journal of Immunology
carbohydrate recognition domain (21). Biosynthesis of SP-D in-
cludes assembly of monomeric SP-D into trimers followed by oli-
gomerization of four trimers via critical cysteine residues at posi-
tions 15 and 20 located within the hydrophobic tails to produce a
dodecamer (23, 24). When fully assembled, SP-D is capable of
mediating a variety of functions, including aggregation of patho-
gens, lysis of microbes, enhancement of phagocytosis, modulation
of cytokines and reactive species, and mediation of effector cell
The mechanistic regulation of the observed pro- and antiinflam-
matory functions of SP-D is incompletely defined. However, based
on data published for SP-A, which has a similar functional duality,
it has been suggested that proinflammatory functions of collectins
are mediated by tail domains, via binding to CD91 and calreticulin,
and antiinflammatory functions by the head domains, via binding
to signal regulatory protein-? (SIRP?). Recent work from our
group has shown that both in vitro and in vivo, SP-D is capable of
undergoing posttranslational modification by NO to produce S-
nitrosylated forms that disrupt higher order oligomerization. Fur-
thermore, S-nitrosylated SP-D (SNO-SP-D) serves as a proinflam-
matory mediator enhancing macrophage migration and lung
chemokine production (25).
Immune reconstitution disease (IRD) in response to a number of
microorganisms has been described in patients immunosuppressed
by HIV infection and by other mechanisms, including chemother-
apy (26–30). In the context of PCP, this syndrome encompasses an
acute symptomatic respiratory decompensation that is related tem-
porally to treatment of PCP coupled with reconstitution of host
immune processes (such as reduction in the dosage of corticoste-
roids and/or cytotoxic agents or a reduction in HIV viral load).
This combination results in the development of immunopathologic
lung damage and acute respiratory failure. Although highly active
antiretroviral therapy has led to a decrease in opportunistic infec-
tions (31), paradoxical worsening of lung function and respiratory
failure have been reported after the beginning of highly active
antiretroviral therapy in patients treated for PCP (26–30). It has
been proposed that this phenomenon results from transient wors-
ening of inflammation due to pulmonary recruitment and activa-
tion of immune cells responding to persistent Pneumocystis cysts
Immmune reconstitution in the setting of PCP has been predom-
inantly modeled in murine hosts using selective transfer of lym-
phocyte subsets. Pneumocystis murina-infected scid mice sub-
jected to direct immunologic reconstitution with selective
populations of sensitized CD4 and/or CD8 T cells mount protec-
tive responses to the organism that results in focal areas of pul-
monary inflammation marked by elaboration of IL-1?, IL-6,
IFN-?, as well as macrophage-derived TNF-? near the sites of
cell-organism contact (9, 32–36). In the present study we have
modified an established and validated model of PCP that utilizes
selective depletion of CD4?cells from mice followed by intratra-
cheal inoculation with P. murina (37–39). To model IRD, with-
drawal of the depleting Ab was performed before the development
of overt PCP (?2 wk). We found that during the ensuing 4 wk,
these reconstitutions produce significant pulmonary inflammatory
responses characterized by depletion and dysfunction of hydro-
phobic pulmonary surfactant components. Additionally, selective
up-regulation of SP-D expression occurs accompanied by alter-
ations in its quaternary structure and function mediated by local
S-nitrosylation of SP-D monomers. These findings represent a
novel paradigm for modulation of lung inflammation by an intrin-
sic innate host defense protein subjected to a selective, nonenzy-
matic posttranslational modification. Furthermore, the results also
extend our understanding of the mechanisms of IRD during PCP
that could suggest novel therapeutic approaches for this increas-
ingly important clinical problem.
Materials and Methods
Monospecific, polyclonal surfactant protein antiserum against SP-B has
been previously characterized in detail (40). A monospecific, polyclonal
Ab against SP-D (AB 1754) was produced in rabbits using synthetic pep-
tides corresponding to two homologous regions of the mouse/human SP-D
sequences as the immunizing Ag and has been previously described (41).
This Ab recognizes denatured isoforms of mouse SP-D as well as both
denatured and native forms of human SP-D. A polyclonal antiserum
against recombinant mouse SP-D was purchased from Chemicon. This
antiserum recognizes multiple isoforms (native greater than denatured) of
Mouse model of Pneumocystis infection and
C57BL/6 mice were purchased from Charles River Laboratories and
housed in a barrier isolation animal care facility at the University of Penn-
sylvania in filter-top cages for 7 days before inoculation. Experiments were
performed between 8 and 14 wk of age on male and female mice. Mice
received sterile rodent chow and sterile drinking water. Normal sentinel
mice were examined routinely for the presence of unintended pathogens by
culture and serology. The Institutional Animal Care and Use Committees
of the University of Pennsylvania reviewed and approved all animal
Organisms. P. murina organisms were obtained from the lungs of athymic
mice (nu/nu on a BALB/c background from Taconic Laboratories) in
which Pneumocystis organisms are propagated by serial passage as previ-
ously described (37). Before dispersal of P. murina by homogenization
using a Stomacher apparatus, bacterial contamination was excluded by the
routine use of Gram staining of touch preparations of each harvested lung.
Following centrifugation, organisms collected in the resulting pellet were
stained with modified Giemsa stain, counted, and then inoculated intratra-
cheally (0.1 ml ? 2 ? 105P. murina cysts) into anesthetized mice.
Generation of P. murina infection and IRD. The experimental design uti-
lized in these studies is illustrated in Fig. 1. C57BL/6 mice were subjected
to selective CD4 depletion via i.p. injection twice weekly with the mAb
GK1.5 as previously published (42–44). Control mice received i.p. injec-
tion of equal volumes of PBS. One week after initiation of CD4 depletion,
mice were immunosuppressed by selective CD4 depletion via i.p. injection
twice weekly with the mAb GK1.5. One week following the initiation of
CD4 depletion, mice were inoculated with P. murina cysts as described in
Materials and Methods. Immunosuppression was continued for 2, 4, and 6
wk to promote P. murina infection as previously published (42). B, IRD
was produced by discontinuation of immunosuppression through with-
drawal of GK1.5 2 wk after inoculation. Mice were then sacrificed 2 or 4
wk (i.e. 2 wk ? 2 wk or 2 wk ? 4 wk groups, respectively) after removal
CD4-depleted mouse model of PCP and IRD. A, C57BL/6
2278IMMUNE RECONSTITUTION AND SURFACTANT IN PCP
experimental animals were intratracheally inoculated under direct visual-
ization with 2 ? 105P. murina cysts harvested from the colony of P.
murina-infected nu/nu mice. Following inoculation, P. murina infection
was allowed to progress for 2 wk before initiation of immune reconstitution
through withdrawal of GK1.5 Ab for an additional 2 or 4 wk as indicated
(i.e., IRD). As positive controls, mice inoculated with P. murina were
subjected to continuous CD4 depletion with GK1.5 for up to 6 wk.
FACS analysis of CD4 cells. Flow cytometric analysis was used to verify
depletion of CD4?T cells at the time of inoculation and to track immune
reconstitution after withdrawal of GK1.5 treatment. At indicated time
points, spleens from nondepleted, depleted, and immune-reconstituted
mice were passed through a wire mesh, and splenocytes were isolated by
gradient centrifugation with Histopaq-1077 (Sigma-Aldrich). Cells (5 ?
105/well) were incubated on ice for 30 min with FITC-labeled anti-CD3
and allophycocyanin-labeled anti-CD4 (clone RM4-5) or isotype-matched
irrelevant Abs (all from BD Pharmingen). Data were collected on a
FACSCalibur flow cytometer and analyses were performed using
CellQuest software (BD Biosciences). Data are expressed as a percent-
age of CD3 cells with positive staining for CD4.
Quantitation of pneumocystis infection. The magnitude of viable organism
burden in P. murina-inoculated mice was assessed by quantitative RT-PCR
using the method of Zheng et al. (45). Total RNA was isolated from the left
lungs of infected mice by a single-step method using TRIzol reagent (In-
vitrogen). Plasmids containing P. murina rRNA (a kind gift from Dr. Chad
Steele, University of Alabama-Birmingham) were used as a standard for
the assay. The template was digested with RNase-free DNase, quantitated
by spectrophotometry, and aliquoted at ?80°C until further use. The Taq-
Man PCR primers for mouse Pneumocystis rRNA are 5?-ATG AGG TGA
AAA GTC GAA AGG G-3? and 5?-TGA TTG TCT CAG ATG AAA AAC
CTC TT-3?. The probe was labeled with a reporter fluorescent dye, FAM,
and the sequence was FAM-AAC AGC CCA GAA TAA TGA ATA AAG
TTC CTC AAT TGT TAC-TAMRA. Real-time PCR was conducted using
sults in clearance of P. murina. PCP in continuously CD4-depleted mice
and IRD mice were generated as described in Materials and Methods and
schematically illustrated in Fig. 1. A, For flow cytometric analysis, total
spleen cells stained for CD3 and CD4 as described in Materials and Meth-
ods were subjected to FACS. The data were expressed as the percentage of
CD4 expressing CD3?T cells (mean ? SEM; n ? 4–8 in each group). At
baseline, nondepleted controls demonstrated 54% CD4?T cells. ?, p ?
0.05 for reconstituted group vs corresponding CD4-depleted group at the
same interval postinfection; #, p ? 0.05 vs from corresponding treatment
group; &, p ? 0.05 vs nondepleted mice. B, P. murina burden after with-
drawal of GK1.5. Viable P. murina were quantitated by real-time PCR
measurement of rRNA copy number using a standard curve of known copy
number of P. murina 18S RNA as described in Materials and Methods.
Data are expressed logarithmically as copy number (mean ? SEM; n ?
4–8 in each group). ?, p ? 0.05 for IRD mice vs corresponding CD4-
depleted group at same time post infection; #, p ? 0.05 vs from corre-
sponding treatment group; @, p ? 0.05 vs CD4-depleted PCP group 2 wk
Recovery of CD4?cells after withdrawal of GK1.5 Ab re-
duces significant worsening of lung inflammation. A, Representative mor-
phological changes in formalin-fixed, paraffin-embedded, H&E-stained
right lung sections prepared from uninfected and P. murina-infected CD4-
depleted or IRD mice harvested 2, 4, and 6 wk postinoculation as labeled.
Original magnification ?100. B, Histological scoring of lung inflamma-
tion. Median inflammation scores were determined by blinded evaluation
of stained sections from each treatment group as described in Materials
and Methods. #, p ? 0.05 vs CD4-depleted 2 wk P. murina infected group.
Immune reconstitution following P. murina infection in-
duces infiltration of lung parenchyma by CD4 and CD8 cells. Total RNA
isolated from the left lungs of mice as in Fig. 2 was reverse transcribed, and
CD4, CD8, and 18S RNA signals were amplified as described in Materials
and Methods. Ct values obtained were normalized to 18S signals and fur-
ther analyzed using the relative quantitation (??Ct) method. Data are ex-
pressed as fold change (mean ? SEM; n ? 5 in each group). ?, p ? 0.05
for IRD mice vs corresponding CD4-depleted group at same time postin-
fection; &, p ? 0.05 vs uninfected and nondepleted group.
Immune reconstitution following P. murina infection in-
2279The Journal of Immunology
one-step TaqMan RT-PCR reagents (Applied Biosystems). The PCR am-
plification was performed for 40 cycles, with each cycle at 94°C for 20 s
and 60°C for 1 min, in triplicate using the ABI Prism 7700 SDS. Threshold
cycle values were averaged triplicate reactions, and data were converted to
rRNA copy number by using a standard curve of known copy number of
P. murina rRNA. This assay has a correlation coefficient ?0.98 over 8 logs
of P. murina rRNA concentration and correlates with viable organism
Histology and inflammation scores. Following lavage, left lungs were re-
moved and frozen in liquid nitrogen for RT-PCR analysis; right lungs were
inflated and fixed with paraformaldehyde (4% in 0.1 M sodium cacodylate
(pH 7.3)) for histological analysis. Paraffin-embedded lung sections stained
with H&E were used to evaluate the intensity of pulmonary inflammation.
Sections were scored in a blinded fashion by two independent observers to
grade the intensity of inflammation using a previously validated scoring
RT-PCR for CD4/CD8 expression. To determine the microenvironment of
the lung parenchyma subjected to IRD, CD4 and CD8 mRNA levels in
harvested lung tissue were quantitated by RT-PCR as described by Phares
et al. (46) and Jassare et al. (47). Total RNA used for determination of P.
murina rRNA described above was reverse transcribed using the TaqMan
reverse transcription reagents kit (Applied Biosystems). CD4 and CD8
were then amplified using the following primers: CD4 (3?-GAG ATT ATG
GCT CTT CTG CAT, 5?-ATC AGG AAG TGA ACC TGG TG) and CD8
(3?-TTC TCT GAA GGT CTG GGC TT, 5?-CAG CAA CTC GGT GAT
GTA CT), a kind gift from Dr. Steven Albelda (University of Pennsylva-
nia). PCR amplification of triplicate cDNA was performed on an Applied
Biosystems 7500 Fast real-time PCR system for 40 cycles, with each cycle
at 95°C for 15 s and 60°C for 1 min. Ct values obtained using Sequence
Detection Software version 1.4 (Applied Biosystems) were determined,
and relative amounts of specific mRNA were calculated using the relative
quantitation (??Ct) method and expressed as fold change. 18S RNA
served as the endogenous control (assayed using a TaqMan gene expres-
sion assay from Applied Biosystems).
Bronchoalveolar lavage fluid (BALF) analyses
Cell counts. Lungs were lavaged with 0.5-ml aliquots of sterile saline to
a total of 5 ml. Recovered BALF samples were centrifuged (400 ? g for
10 min) and the cell pellet was gently resuspended in 1 ml of PBS (with
Ca2?and Mg2?) for total cell count determination using a Z1 Counter
particle counter (Beckman Coulter). Aliquots of cells were spun on a
Thermo Shandon Cytospin-3 at 750 rpm for 3 min and stained with stan-
dard Diff-Quik for manual determination of cell differentials. Cells were
identified as macrophages, eosinophils, neutrophils, and lymphocytes by
increased BALF cellularity. A, Total BALF cell counts and (B) differential
cell counts for macrophages (MP), lymphocytes (LC), eosinophils (EO),
and neutrophils (NP) were performed as described in Materials and Meth-
ods in P. murina-infected CD4-depleted or IRD mice harvested 2, 4, and
6 wk postinoculation. The data are expressed as cell numbers ? 1000.
Values are shown as mean ? SEM (n ? 5–20 animals in each group).
Multiple comparisons were made by ANOVA. ?, p ? 0.05 for IRD mice
vs corresponding CD4-depleted group at same time postinfection; #, p ?
0.05 vs corresponding treatment group; ∧, p ? 0.05 vs uninfected group.
P. murina in immune-reconstituted mice is associated with
fection is enhanced in IRD mice. A, Total
protein content of BALF fractions of unin-
fected and P. murina-infected CD4-depleted
or IRD mice was determined as described in
Materials and Methods. Data are expressed
as mean ? SEM (?g) per mouse. B, Body
weights and lung weights from uninfected
and P. murina-infected CD4-depleted or
IRD mice were recorded. Group mean data
(?SEM) are expressed as a ratio of lung-to-
body weight at terminal endpoints. C and D,
Quantitation of total NO (C) and nitrates (D)
in BALF of uninfected and P. murina-in-
fected CD4-depleted or IRD mice. BALF
samples were analyzed by chemical reduc-
tion and chemiluminesence as described in
Materials and Methods. Data are expressed
as mean ? SEM (in nmol); n ? 4–8 in each
group. For all panels, ?, p ? 0.05 for IRD
mice vs corresponding CD4-depleted group
at same time postinfection; #, p ? 0.05 vs
corresponding treatment group; ∧, p ? 0.05
vs uninfected group.
Lung injury by P. murina in-
2280IMMUNE RECONSTITUTION AND SURFACTANT IN PCP
Analysis of surfactant components. Cell-free BALF supernatants were
separated into large-aggregate (LA) and small-aggregate (SA) fractions by
centrifugation (20,000 ? g for 60 min at 4°C) as described previously (7).
Total protein content of LA and SA fractions was determined by the
method of Bradford, with bovine IgG as a standard (48). Total phospho-
lipid content of LA and SA fractions was determined by the method of
Surface tension measurements. The biophysical activity of recovered sur-
factant from all experimental groups was measured in a capillary surfac-
tometer (Calmia Medical) as described in detail previously (7, 50, 51).
Briefly, LA fractions of BALF were diluted with saline to a total phos-
pholipid concentration of 1 mg/ml and 0.5-?l samples and then introduced
into the glass capillary of the capillary surfactometer and compressed for
120 s, resulting in cyclic extrusion from the narrow end of the capillary
permitting airflow during capillary patency. The percentage of the 120-s
study period that the capillary is patent was calculated and data were ex-
pressed as percentage openness. Each sample was analyzed in triplicate.
PAGE and immunoblotting. BALF proteins were separated and analyzed
by two methods. Denaturing SDS-PAGE was performed under reducing
conditions using 10–20% Novex tricine gels for SP-B and NuPAGE Novex
10% Bis-Tris gels for SP-D (all from Invitrogen).
Native gel electrophoresis for detection of SP-D quaternary structure
was performed using NuPAGE 3–8% Tris-acetate (Invitrogen) as previ-
ously described (25). Calculated equal amounts of SP-D determined first
using SDS-PAGE as described above were mixed with a cold native Tris-
glycine sample buffer before loading. Electrophoresis was run at room
temperature at a constant voltage of 150 V for 2 h. Proteins were then
transferred to polyvinylidene difluoride membranes.
Separated proteins (1 ?g of total protein per lane) were transferred to
nitrocellulose at room temperature using Tris-glycine transfer buffer at 30
V overnight. Blots were blocked for 1 h at room temperature with 10%
nonfat milk and then incubated with primary SP-B (1/5000 dilutions) or
SP-D Ab (1/20,000 dilution) for 1 h. The intensity of bands visualized
using HRP-conjugated goat anti-rabbit IgG (Jackson ImmunoResearch
Laboratories) and ECL (Amersham Biosciences) were quantitated by den-
sitometric scanning of exposed films or direct acquisition on a Kodak 440
Biotin switch assay for detection of SNO-SP-D. Detection of SNO-SP-D
was performed via an adaptation of the biotin switch method (52). BALF
(30 ?g of total protein) was diluted in HEN buffer (25 mM HEPES (pH
7.7), 0.1 mM EDTA, 0.01 mM neocuproine) and 20 ?M N-ethylmaleimide
at 37°C for 30 min to block free thiols. Excess N-ethylmaleimide was
removed by protein precipitation using cold acetone. Protein pellets were
resuspended in diethylenetriaminepentaacetic acid and S-nitrosothiol
(SNO) bonds decomposed by adding 20 mM sodium ascorbate. The newly
formed thiols were linked with the sulfhydryl-specific biotinylating reagent
Pierce Biotechnology). Biotinylated proteins were precipitated with
streptavidin-agarose beads, and Western blot analysis was performed to
detect the amount of captured SP-D using polyclonal SP-D antiserum.
NO measurements. BALF samples were analyzed for NO metabolites by
chemical reduction and chemiluminescence using the Ionics/Sievers nitric
oxide analyzer 280 (NOA 280; Ionics Instruments), as previously described
(53). All nitrogen oxides were reduced by use of an excess of vanadium
chloride in hydrochloric acid at 95°C, and measurements using these con-
ditions were considered as a total nitrogen oxide measurement.
Nitrite was measured independently and subtracted from the total NO to
calculate the concentration of nitrate. Nitrite analysis was performed using
tion with allogenenic lung homogenate does not produce significant lung
injury. CD4-depleted C57BL/6 mice were inoculated with 100 ?l of lung
homogenate (LH) from uninfected nu/nu (BALB/c background) donors. A
control group consisted of mice subjected to CD4 depletion only for 6 wk
(CD4 Depleted). Four weeks postinoculation, GK1.5 was withheld from
indicated groups (4 ? 2; 4 ? 4) to produce immunoreconstitution. Mice
were sacrificed as indicated at 4, 6, and 8 wk after LH inoculation. A, Total
BALF cell counts were determined using a Coulter counter as described in
Materials and Methods. Data are expressed as cell numbers ? 1000. Val-
ues are shown as mean ? SEM (n ? 5–7 animals in each group). B, Total
protein content of BALF fractions was determined by the method of Brad-
ford (48). Data are expressed as mean ? SEM (?g) per mouse.
Immunoreconstitution of C57BL/6 mice following inocula-
Table I. BALF cytokine levels in Pneumocystis infection and IRDa
Pneumocystis Infected (Weeks Postinoculation)
2462 ? 22 ? 4
72.1 ? 15.4
10.6 ? 9.0
19.2 ? 3.7
25.0 ? 6.8
17.1 ? 8.7
96.6 ? 21.1
15.8 ? 7.4
37.8 ? 7.4
27.8 ? 7.7
20.6 ? 4.3
192.1 ? 48.4
74.2 ? 32.1
99.1 ? 22.0
47.6 ? 7.8
56.1 ? 11.5
231.9 ? 42.1*
143.6 ? 70.9*
92.3 ? 24.1
46.3 ? 11.9
122.2 ? 11.2*
220.3 ? 48.3*
97.1 ? 27.3
149.8 ? 28.9*
54.7 ? 19.0
67.2 ? 3.6
256.2 ? 74.7*
68.8 ? 35.6
169.1 ? 46.3*
41.8 ? 9.7
192.8 ? 38.1*,#
aMultiplex analysis of BALF for all groups for cytokine/chemokine was performed as described in Materials and Methods. Data are expressed as total cytokine recovered
(in pg) as mean ? SEM (n ? 5 in each group). Groups were compared using one-way ANOVA. ?, p ? 0.05 vs uninfected (Lung homogenate inoculated) mice; #, p ? 0.05
vs 6 wk CD4-depleted Pneumocystis-infected group.
2281The Journal of Immunology
a KI and acetic acid mixture at room temperature (54). Resultant signal
areas from each assay were compared with standards to calculate the con-
centration of each nitrogen oxide. Sodium nitrate and nitrite (Sigma-
Aldrich) were utilized as the standards for the vanadium and iodide assays,
Measurement of cytokines. Sandwich ELISA assays of BALF from all
groups for levels of MCP-1 (CCL2), eotaxin, and KC were performed
using Quantikine kits from R&D Systems following the manufacturer’s
instructions. BALF IFN-? and TNF-? were measured using sandwich
ELISA kits from BD Pharmingen. All samples were assayed in duplicate.
Chemotaxis assay. Directed migration (chemotaxis) of cells was per-
formed as previously described (25). Briefly, 50 ?l of RAW 264.7 cells
(American Type Culture Collection), suspended at 2 ? 106cells per ml in
DMEM, was placed in the upper wells of a 48-well microchemotaxis
chamber (Neuro Probe). The lower chambers contained 40 ?l of test so-
lution, consisting of DMEM and either saline (control) or BALF from PCP
infected or reconstituted mice. A polyvinylpyrrolidone-free polycarbonate
filter (5-?m pores) was placed between the wells along with the rubber
gasket of the assembly. The chamber was incubated for 3 h at 37°C with
5% CO2. Nonmigrating cells were scraped from the upper surface, and the
filter containing migrating cells was stained with Hemacolor differential
blood stain and mounted on a glass coverslip. Cells migrated through the
filter were counted in 10 randomly selected oil-immersion fields in each
well at ?100 magnification. Data were expressed as the average of the
three fields in cells per oil-immersion field.
Data analyses were performed using GraphPad InStat v3.06 for Windows
(GraphPad Software). Parametric data were analyzed with ANOVA or Stu-
dent’s t test assuming equal variances to test differences between groups.
Data were expressed as mean ? SEM. Nonparametric data were analyzed
by the Wilcoxon/Kruskal-Wallis rank sum test. Data were expressed as
median values. In all cases a p value of ?0.05 was considered as
Kinetics of P. murina infection and CD4 cell recovery in a
model of IRD
Administration of two doses of GK1.5 mAb resulted in complete
depletion of peripheral CD4?T cells (Fig. 2A). Following with-
drawal of the Ab 2 wk after intratracheal inoculation of P. murina,
CD4 T cells in the periphery recovered over the ensuing 4 wk,
reaching 65% of nondepleted mouse levels.
In mice persistently depleted of CD4 cells, progressive infection
with P. murina occurred. Using a sensitive and quantitative RT-
PCR protocol, P. murina-specific rRNA was detected at 2 wk
postinoculation and increased progressively to week 6 (Fig. 2B). In
contrast, the P. murina burden in mice undergoing immune recon-
stitution at week 2 had similar degrees of viable organisms 4 wk
after inoculation. However, following the return of significant
numbers of peripheral CD4 T cells (4 wk), there was a significant
reduction in P. murina burden in the IRD group.
Quantification of lung inflammation and cellular accumulation
Despite significantly lower organism burdens, IRD mice infected
with P. murina developed equivalent amounts of parenchymal pul-
monary inflammation (Fig. 3A). Scoring of histopathology from
these groups demonstrated progressive cellular inflammation in the
CD4-depleted group through 6 wk postinoculation, which was
similar to that observed in the IRD group even 4 wk after GK1.5
withdrawal (Fig. 3B). Furthermore, using RT-PCR to detect ex-
pression of T cell surface Ags (Fig. 4), the lung parenchyma from
the IRD group was found to be infiltrated with a combination of
CD4 and CD8 cells while, as expected and consistent with previ-
ous reports (55), lung tissue from the CD4-depleted group infected
with P. murina had marked increases in CD8 expressing cells.
The increase in parenchymal cellular infiltrates in IRD mice was
accompanied by a commensurate increase in total BALF cell
counts (Fig. 5A). Diff-Quik staining of cytospins from the BALF
At the indicated time postinoculation, the biophysically active LA surfactant
fraction was prepared from harvested BALF. A, Samples of LA fractions were
separated by SDS-PAGE and immunoblotted with SP-B Ab as described in
Materials and Methods. Band density was quantified and is expressed as per-
centage of uninfected level (mean ? SEM; n ? 5 in each group). B, Total
phospholipid in LA was estimated using a modification of the colorimetric
Bartlett method as described in Materials and Methods. Data are expressed as
was determined by measuring capillary openness by capillary surfactometer as
described in Materials and Methods. Values are obtained by averaging tripli-
cate measurements of each sample and group mean data (mean ? SEM, ex-
pressed as percentage of capillary openness (100 being fully open); n ? 4–6
samples/time point). For all panels, ? p ? 0.05 for IRD mice vs corresponding
CD4-depleted group at same time postinfection; #, p ? 0.05 vs corresponding
treatment group; ∧, p ? 0.05 vs uninfected group.
Surfactant biophysics and function are impaired in IRD mice.
2282IMMUNE RECONSTITUTION AND SURFACTANT IN PCP
subjected to differential cell counting demonstrated that the alve-
olar cell population consisted predominantly of macrophages but
also had a significant degree of neutrophilia and increased numbers
of lymphocytes (Fig. 5B).
Measurement of lung injury and alteration in surfactant
Importantly, despite a ?95% reduction in P. murina organism bur-
den, at 6 wk postinoculation (and 4 wk postreconstitution), IRD
mice continued to display elevated amounts of total BALF protein
similar in magnitude to persistently infected mice (Fig. 6A). This
finding was also reflected in increased lung-to-body weight ratios
in both groups (Fig. 6B). Taken together, the indices of lung injury
following IRD are enhanced relative to the degree of infection.
These changes in cellularity and protein leak were not seen when
mice were CD4 depleted, inoculated with lung homogenate from
uninfected nu/nu (BALB/c background) donors, and then sub-
jected to immune reconstitution, indicating that the inflammatory
responses were not attributable to the use of allogenic donor mice
Inflammatory mediators present in the BALF of the injured
groups were also assessed. Both P. murina-infected and IRD mice
demonstrated elevations in total BALF NO and its oxidative forms
(nitrates) (Fig. 6, C and D). Multiplex BALF cytokine analysis
revealed that 6 wk after inoculation, there were marked elevations
in MCP-1 in both P. murina-infected (CD4 depleted) mice as well
as IRD (“2 ? 4”) mice, although the increases were slightly greater
in the IRD group (Table I). Compared with uninfected controls,
IRD mice had significant increases in KC and IFN-?, while P.
murina-infected (6 wk) groups had significant elevations in IFN-?,
but there were no differences between the two groups.
Biochemical analysis of BALF revealed that 6 wk postinocula-
tion, despite the marked variations in organism burden, CD4-de-
pleted and IRD mice each had marked decreases in SP-B levels in
the biophysically active large aggregate surfactant fraction (Fig.
8A), which was accompanied by similar decreases in total phos-
pholipid (Fig. 8B). The alterations in SP-B protein and phospho-
lipid contents were reflected in a dysfunctional surfactant in both
groups where, using a capillary surfactometer, moderate reduc-
tions in surface activity could be seen (Fig. 8C). Consistent with
the results obtained for cell counts and BALF protein (Fig. 7), total
phospholipid in large aggregate surfactant fractions was not altered
in mice receiving intratracheal uninfected BALB/c lung homoge-
nate followed by withdrawal of GK1.5 (data not shown).
IRD promotes nitrosylation of SP-D
Despite decreases in SP-B and phospholipid that evolved during
immune reconstitution, SP-D levels were significantly increased in
mice subjected to immune reconstitution within 2 wk after with-
drawal of GK1.5. By Western blotting and quantization, IRD mice
had SP-D levels that were 3-fold higher than uninfected controls
and 60% greater than mice continuously infected with P. murina
(Fig. 9A). Recently, we have shown that inflammatory lung injury
can induce alterations in SP-D structure and function through S-
nitrosylation of cysteine residues in its NH2tail region (25). We
therefore subjected BALF from these mice to analysis for the pres-
ence of SNO-modified SP-D and performed native gel electro-
phoresis to determine molecular subspecies of SP-D. As shown in
Fig. 9B, using a biotin derivitization method, SP-D in the BALF of
immunoreconstituted mice showed increased levels of S-nitrosy-
lation that were accompanied by marked alterations in quaternary
structure. Under these electrophoresis conditions, native SP-D
from CD4-depleted mice with Pneumocystis infection is too large
(Mr? 800,000) to enter the 10% resolving gel and it remains in the
of SNO-SP-D. A, Total SP-D in BALF was determined by Western blotting
and densitometric scanning as described in Materials and Methods. Data
were normalized to uninfected control and are reported as mean ? SEM
(expressed as percentage of control; n ? 4–6 samples per time point).
?, p ? 0.05 for IRD mice vs corresponding CD4-depleted group at same
time postinfection; #, p ? 0.05 vs corresponding treatment group; ∧, p ?
0.05 vs uninfected group. B, BALF from P. murina-infected CD4-depleted
or IRD mice 6 wk postinoculation was analyzed for SNO-SP-D content by
the biotin switch assay as described in Materials and Methods. Shown are
blots obtained from two separate analyses of two independent experiments.
SNO-SP-D formation is consistently increased in the IRD samples. C,
BALF normalized for equal calculated amounts of total SP-D (as deter-
mined by reduced SDS PAGE) in A from P. murina-infected CD4-depleted
or IRD mice 6 wk postinoculation was analyzed by native PAGE and
Western blotting with SP-D antiserum to determine quaternary structure of
SP-D. Multimeric SP-D is incapable of entering the gel at the top, while
smaller molecular mass forms are seen only in BALF from reconstituted
mice. Data are representative of duplicate determinations from two inde-
IRD alters the quaternary structure of SP-D and production
2283The Journal of Immunology
wells (Fig. 9C). Similar patterns were obtained for BALF from
uninfected mice (data not shown). In contrast, BALF from IRD
mice that contained marked amounts of S-nitrosylated SP-D ex-
hibited significant amounts of smaller molecular forms (trimers).
Taken together, these data indicate that a selective and specific
modification of SP-D by NO occurs during the inflammatory re-
sponse associated with IRD.
We have recently shown that S-nitrosylated SP-D could alter the
chemotactic ability of macrophages (25). Both P. murina infection
and IRD produced increases in alveolar macrophage numbers 6 wk
after Pneumocystis inoculation. Based on this, we hypothesized
that under conditions of IRD, the alterations in SP-D structure we
observed could functionally promote macrophage chemotaxis. We
examined the ability of both modified and unmodified SP-D to act
as a chemoattractant for macrophages. Utilizing a modified Boy-
den chamber with RAW cells as the target, BALF from IRD mice
promoted significantly greater chemotaxis than did that from CD4-
depleted animals (Fig. 10). Pretreatment of the BALF with ascor-
bate, which removes the NO moiety from SNO, significantly re-
duced the chemotactic efficacy of BALF to levels similar to those
seen with PCP and CD4 depletion.
The coordinated regulation of the immune response to promote
organism clearance and then limit local tissue damage is crucial
to effective lung host defense. Multiple studies have shown that
CD4 T cells are essential for proper clearance of Pneumocystis
pulmonary infection (36, 38, 56). Clinically, several case re-
ports and series have detailed PCP patients who have undergone
immune recovery and developed acute respiratory failure (26–
29). We have developed a murine model of IRD in which the
return of native CD4?T cells through the withdrawal of an im-
munosuppressive mAb, GK1.5, mimics the clinical disease in hu-
mans. In this model, as in humans with IRD, we observed marked
increases in pulmonary inflammation and lung injury parameters
despite significantly attenuated P. murina burden. Utilizing this
model, the present study extends previous observations of IRD
pathogenesis through identification of a novel form of SP-D post-
translationally modified by S-nitrosylation that profoundly alters
its normal immunosuppressive effects on lung effector cell
It has been previously well documented that in the pathogenesis
of respiratory failure in PCP, the surfactant system plays important
role in the modulation of lung mechanics and of gas exchange
(7–9). In the present study, we documented changes in surfactant
component expression and biophysical activity, including selective
down-regulation of phospholipid and SP-B along with correspond-
ing increases in surface tension. However, immune reconstitution
resulted in an equally pronounced surfactant dysfunction, suggest-
ing that the host response to residual Pneumocystis organisms or
Ags can further modulate inhibition of surfactant activity. Coupled
with the finding that uninfected lung homogenate failed to generate
either a significant inflammatory injury (Fig. 7) or disruption of
surfactant component expression, anti-Pneumocystis responses
that occurred during CD4-mediated immune recovery were re-
sponsible for the observed enhancement of lung injury in mice and
could be similarly responsible for the significant morbidity ob-
served in patients with PCP and IRD.
While Pneumocystis organisms have been shown to attach di-
rectly to alveolar epithelial cells (57), it is unlikely that Pneumo-
cystis is playing a direct role in the IRD-mediated lung damage. In
vitro we have shown that P. murina does not alter alveolar epi-
thelial cell barrier function (58), indicating that it is more likely
that local inflammatory effector cells are mediating the observed
lung damage. Furthermore, in the present study, equivalent or
greater injury occurred despite decreased burdens of P. murina.
The lungs of IRD mice showed increased lung edema and lung-
to-body weight ratios. Previously, other investigators have utilized
reconstitution with bone marrow-derived CD4 and CD8 cells to
model Pneumocystis-induced IRD in scid/scid mice (33, 34, 36,
59). Under normal circumstances, this strain typically develops
low levels of pulmonary inflammation despite high organism bur-
dens. When reconstituted with CD4 cells, those mice developed
marked increases in lung injury and gross physiological changes
through SNO-SP-D. A, BALF from uninfected or P. murina-infected
CD4-depleted mice with or without IRD was harvested 2, 4, or 6 wk
after inoculation and assayed for the ability to induce RAW 264.7 mac-
rophage migration using a modified Boyden chamber. Migration, de-
fined as the number of cells transitioning the barrier membrane after 3 h
of incubation, was determined by manual counting. Data represent
group mean values (?SEM) from measurements performed in triplicate
from two independent experiments and analyzed by ANOVA. ?, p ?
0.05 for IRD mice vs corresponding CD4-depleted group at same time
postinoculation; #, p ? 0.05 vs corresponding treatment group; ∧, p ?
0.05 vs uninfected group. B, To eliminate the effect of SNO modifica-
tion, BALF from P. murina-infected CD4-depleted mice or IRD mice
was harvested 6 wk after inoculation and pretreated with 20 mM ascor-
bic acid or PBS as indicated and analyzed for the ability to induce RAW
cell chemotaxis in vitro as in A. Data are normalized as percentage of
P. murina-infected continuously CD4-depleted mice (6 wk). All mea-
surements were performed in triplicate and are representative of two
independent experiments analyzed by ANOVA. ?, p ? 0.05 for PBS
treated BALF from IRD mice vs corresponding CD4-depleted group;
#, p ? 0.05 vs corresponding PBS-treated BAL.
Immune-reconstitution induces macrophage chemotaxis
2284IMMUNE RECONSTITUTION AND SURFACTANT IN PCP
associated with decreased lung compliance. Although the molec-
ular mechanisms underlying the observed findings were not com-
pletely defined, the time course (4–6 wk) for lung injury seen in
those studies appears to be similar to the model utilized for the
In this model, histological scores of parenchymal inflammation
were significantly increased in both P. murina-infected and IRD
groups (Fig. 3B). However, using more sensitive methods, we
were able to detect subtle differences in the cellular composition of
the inflammatory response that were consistent with immune re-
constitution. By RT-PCR, the infiltration of the lung parenchyma
that occurred during IRD consisted of both CD4 and CD8 express-
ing cells, while responses in CD4-depleted (GK1.5-treated) mice
infected with P. murina were limited to CD8 cells (Fig. 5). Fur-
thermore, in addition to alterations in lung parenchymal infiltra-
tion, there was a relatively greater increase in BALF cells late in
the course in IRD mice compared with their P. murina-infected
CD4-depleted counterparts (Fig. 4). The histology and accompa-
nying scoring reflect primarily parenchymal inflammation from
accumulation of inflammatory cells in the tissue. In contrast, cells
recovered in the BALF reflect effector cells that have traversed into
the airspaces. The observed dichotomy between parenchyma and
alveolus provides support for the concept that soluble mediators/
chemoattractants preferentially compartmentalized in the airspaces
could promote chemotaxis.
Mechanistically, in addition to increased local effector cells, the
enhanced inflammation and tissue damage seen in the lungs of
mice with IRD appear to be mediated in part by reactive oxygen/
nitrogen species. In previous work in a CD4-depleted model, we
demonstrated elevations in the level of total NO production and
inducible NO synthase (iNOS) protein expression in the BALF cell
pellet during PCP (50, 60). In this study, the BALF from IRD,
Pneumocystis-infected mice contained increased levels of both to-
tal NO and nitrite. We have previously shown both in PCP and in
bleomycin-induced lung injury that enhanced NO/nitrite correlate
with levels of 3-nitrotyrosine, a marker of oxidative-nitrative stress
arising from the reactive product of NO and superoxide (50, 53,
61). Taken together, the data are consistent with the concept that
despite clearance of the organism, NO and its metabolites pro-
duced locally rapidly interact with molecular targets in the lung to
promote damage. The present data also raise the possibility of
using selective inhibition of iNOS as a therapeutic strategy to limit
lung damage from NO-mediated inflammation that occurs during
IRD. A similar approach of iNOS modulation has been shown to
be effective in limiting damage in bleomycin models of injury (62).
In addition to direct tissue damage by reactive oxygen/nitrogen
species, posttranslational modification of selected protein targets
by NO could also occur. One such target of NO appears to be
SP-D. It is well known that the lung collectins can modulate both
pro- and antiinflammatory effects in vitro and in vivo (50, 53,
63–69). Recently, mechanistic insights into this conundrum have
been reported. Using SP-A as the model, the immunomodulatory
effects of lung collectins on macrophage function can be attributed
in part to selective engagement of two different cell surface recep-
tors capable of interacting with distinct spatial domains in the car-
bohydrate recognition head domain and N-terminal tail region
present in SP-A and SP-D (70).
We have recently extended this concept by demonstrating that
the dichotomous functional nature of SP-D can be attributed to
alterations in its multimeric structure by NO (25). In its native
multimeric form, SP-D is antiinflammatory, down-regulating the
inflammatory responses of effector cells. However, in vitro, the
transnitrosylation of SP-D with S-nitrosocysteine produces S-ni-
trosylation of two key cysteine residues (Cys15/Cys20), resulting in
disruption of normal quaternary structure (dodecamer) to produce
predominantly monomeric and trimeric SNO-SPD that is chemoat-
tractive for RAW macrophages and induces p38 MAPK phosphor-
ylation mediated by binding to calreticulin/CD91. This observation
led us to hypothesize that S-nitrosylation of SP-D was the molec-
ular basis for the enhanced lung injury seen in IRD associated with
PCP. Multiple findings documented in the present work support
this concept. First, the increase in lung inflammation observed in
IRD mice was accompanied by increases in alveolar levels of
SP-D. In previous models of P. murina infection, increases in
SP-D occurred late in the course of pulmonary infection (8, 71).
During IRD, P. murina-infected mice subjected to withdrawal of
GK1.5 had even greater increases in total SP-D protein, but there
was an associated marked alteration in its higher order structure.
Second, S-nitrosylation of SP-D was observed in IRD. Third, faster
migrating forms of SP-D were seen in native gel electrophoresis
corresponding to these monomeric and trimeric forms (25) and
were exclusively and markedly enhanced by immunoreconstitution
and not P. murina infection alone (Fig. 9). Fourth, treatment of
BALF with ascorbic acid, which, based on its redox potential,
selectively reduces S-nitrosylated proteins to release NO (52),
blocked the chemotactic activity of BALF obtained from IRD mice
The modification of SP-D by inflammation during IRD repre-
sents an important emerging paradigm in local immunoregulation
in the lung and extends previous similar biochemical and func-
tional observations in a noninfectious mouse model of lung injury.
In two previous studies, we have shown that administration of
bleomycin to mice produces large amounts of macrophage-driven
pulmonary inflammation that was also associated with formation
of S-nitrosylated SP-D (25, 53). Furthermore, treatment with either
anti-SP-D or ascorbic acid blocks macrophage chemotaxis in vitro
(25). Similarly, treatment of IRD BALF with ascorbate in the
present study also blocked chemotaxis. The effect of ascorbate on
macrophage chemotaxis is unlikely to be due to direct effects of
ascorbate on macrophages. We have shown that ascorbic acid is
impermeable to many biological membranes (72), so significant
alterations in intracellular concentrations are not likely. Addition-
ally, the control BALF (6 wk of PCP) in Fig. 10B was also treated
with ascorbate and there was no effect on baseline chemotaxis.
Therefore, it seems unlikely that there is either a significant effect
on macrophages directly (apart from that of SNO-SP-D) or a mod-
ification of other components of the BALF by ascorbate treatment
sufficient to alter chemotaxis. Finally, we have previously shown
that in vitro S-nitrosylation of BALF from SP-D knockout mice
using the SNO donor S-nitrosocysteine has no effect on p38 phos-
phorylation in RAW macrophages, indicating that NO produced
during inflammation is exclusively targeting SP-D and not other
protein components (e.g., SP-A, albumin) of BALF. Taken to-
gether, although chemokines such as MCP-1 are elevated during
PCP and IRD (Ref. 73 and this study), the chemotactic activity of
BALF from IRD mice is almost exclusively attributable to
In conclusion, IRD in a mouse model using withdrawal of
GK1.5 is associated with enhanced pulmonary injury and proin-
flammatory events. In parallel, IRD-mediated lung injury was as-
sociated with marked increases in BALF levels of SP-D. In vitro,
the BALF showed enhanced nitrosylation of SP-D and promoted
increased chemotaxis in a macrophage cell line in vitro. Thus,
inflammation during IRD represents a feed-forward system in
which the additional inflammation leads to further modification of
SP-D by NO and subsequent proinflammatory effects mediated by
these smaller SP-D forms. This study thus emphasizes the delicate
balance that exists between collectins, the innate immune system,
2285The Journal of Immunology
and pulmonary inflammation. Therapeutically, it is possible that
the balance between modified and unmodified SP-D could modu-
late the relative degree of inflammation observed in lung injury,
leading to the potential therapeutic use of native, multimeric SP-
D-containing surfactants. Furthermore, as an alternative strategy,
the local inhibition of NO production may in fact lead to decreased
posttranslational modification of SP-D and limit the feed-forward
inflammatory response and lung damage.
We thank Helchem Kadire for expert technical assistance in the generation
and maintenance of the murine models used in these studies.
The authors have no financial conflicts of interest.
1. Stansell, J. D., D. H. Osmond, E. Charlebois, L. LaVange, J. M. Wallace,
B. V. Alexander, J. Glassroth, P. A. Kvale, M. J. Rosen, L. B. Reichman, et al.
1997. Predictors of Pneumocystis carinii pneumonia in HIV-infected persons:
Pulmonary Complications of HIV Infection Study Group. Am. J. Respir. Crit.
Care Med. 155: 60–66.
2. Su, T. H., and W. J. Martin, 2nd. 1994. Pathogenesis and host response in Pneu-
mocystis carinii pneumonia. Annu. Rev. Med. 45: 261–272.
3. Kovacs, J. A., J. W. Hiemenz, A. M. Macher, D. Stover, H. W. Murray,
J. Shelhamer, H. C. Lane, C. Urmacher, C. Honig, and D. L. Longo. 1984.
Pneumocystis carinii pneumonia: a comparison between patients with the ac-
quired immunodeficiency syndrome and patients with other immunodeficiencies.
Ann. Intern. Med. 100: 663–671.
4. Huang, L., A. Morris, A. H. Limper, J. M. Beck, and on behalf of the ATS
Pneumocystis Workshop Participants. 2006. An Official ATS Workshop Sum-
mary: recent advances and future directions in Pneumocystis pneumonia (PCP).
Proc. Am. Thorac. Soc. 3: 655–664.
5. Kovacs, J. A., V. J. Gill, S. Meshnick, and H. Masur. 2001. New insights into
transmission, diagnosis, and drug treatment of Pneumocystis carinii pneumonia.
J. Am. Med. Assoc. 286: 2450–2460.
6. Beers, M. F., E. N. Atochina, A. M. Preston, and J. M. Beck. 1999. Inhibition of
lung surfactant protein B expression during Pneumocystis carinii pneumonia in
mice. J. Lab. Clin. Med. 133: 423–433.
7. Atochina, E. N., M. F. Beers, S. T. Scanlon, A. M. Preston, and J. M. Beck. 2000.
P. carinii induces selective alterations in component expression and biophysical
activity of lung surfactant. Am. J. Physiol. 278: L599–L609.
8. Atochina, E. N., J. M. Beck, S. T. Scanlon, A. M. Preston, and M. F. Beers. 2001.
Pneumocystis carinii pneumonia alters expression and distribution of lung col-
lectins SP-A and SP-D. J. Lab. Clin. Med. 137: 429–439.
9. Wright, T. W., R. H. Notter, Z. D. Wang, A. G. Harmsen, and F. Gigliotti. 2001.
Pulmonary inflammation disrupts surfactant function during Pneumocystis carinii
pneumonia. Infect. Immun. 69: 758–764.
10. Wiedmann, H. P. 1996. Surfactant and acquired lung diseases. J. Lab. Clin. Med.
11. Martin, W. J., and J. R. Wright. 1999. Pneumocystis carinii pneumonia and
pulmonary surfactant. J. Lab. Clin. Med. 133: 406–407.
12. Griese, M. 1999. Pulmonary surfactant in health and human lung diseases: state
of the art. Eur. Respir. J. 13: 1455–1476.
13. Rooney, S. A., S. L. Young, and C. R. Mendelson. 1994. Molecular and cellular
processing of lung surfactant. FASEB J. 8: 957–967.
14. Crouch, E. C. 1998. Collectins and pulmonary host defense. Am. J. Respir. Cell
Mol. Biol. 19: 177–201.
15. Possmayer, F. 1988. A proposed nomenclature for pulmonary surfactant-associ-
ated proteins. Am. Rev. Respir. Dis. 138: 990–998.
16. Curstedt, T., H. Jornvall, B. Robertson, T. Bergman, and P. Berggren. 1987. Two
hydrophobic low-molecular-mass protein fractions of pulmonary surfactant: char-
acterization and biophysical activity. Eur. J. Biochem. 168: 255–262.
17. Weaver, T. E., C. L. Na, and M. Stahlman. 2002. Biogenesis of lamellar bodies,
lysosome-related organelles involved in storage and secretion of pulmonary sur-
factant. Semin. Cell Dev. Biol. 13: 263–270.
18. Crouch, E. C., and J. R. Wright. 2001. Surfactant proteins A and D and pulmo-
nary host defense. Annu. Rev. Physiol. 63: 521–554.
19. Crouch, E., K. Hartshorn, and I. Ofek. 2000. Collectins and pulmonary innate
immunity. Immunol. Rev. 173: 52–65.
20. Wright, J. R. 2005. Immunoregulatory functions of surfactant proteins. Nat. Rev.
Immunol. 5: 58–68.
21. Crouch, E., A. Persson, D. Chang, and J. Heuser. 1994. Molecular structure of
pulmonary surfactant protein D (SP-D). J. Biol. Chem. 269: 17311–17319.
22. Crouch, E., D. Parghi, S. F. Kuan, and A. Persson. 1992. Surfactant protein D:
subcellular localization in nonciliated bronchiolar epithelial cells. Am. J. Physiol.
23. Zhang, P. N., A. McAlinden, S. Li, T. Schumacher, H. L. Wang, S. S. Hu,
L. Sandell, and E. Crouch. 2001. The amino-terminal heptad repeats of the
coiled-coil neck domain of pulmonary surfactant protein D are necessary for the
assembly of trimeric subunits and dodecamers. J. Biol. Chem. 276: 19862–19870.
24. Brown-Augsburger, P., D. Chang, K. Rust, and E. C. Crouch. 1996. Biosynthesis
of surfactant protein D: contributions of conserved NH2-terminal cysteine resi-
dues and collagen helix formation to assembly and secretion. J. Biol. Chem. 271:
25. Guo, C. J., E. N. Atochina-Vasserman, E. Abramova, J. P. Foley, A. Zaman,
E. Crouch, M. F. Beers, R. C. Savani, and A. J. Gow. 2008. S-nitrosylation of
surfactant protein-D controls inflammatory function. PLoS Biol. 6: e266.
26. Barry, S. M., M. C. I. Lipman, A. R. Deery, M. A. Johnson, and G. Janossy. 2002.
Immune reconstitution pneumonitis following Pneumocystis carinii pneumonia
in HIV-infected subjects. HIV Med. 3: 207–211.
27. Koval, C. E., F. Gigliotti, D. Nevins, and L. M. Demeter. 2002. Immune recon-
stitution syndrome after successful treatment of Pneumocystis carinii pneumonia
in a man with human immunodeficiency virus type 1 infection. Clin. Infect. Dis.
28. Wislez, M., E. Bergot, M. Antoine, A. Parrot, M. F. Carette, C. Mayaud, and
J. Cadranel. 2001. Acute respiratory failure following HAART introduction in
patients treated for Pneumocystis carinii pneumonia. Am. J. Respir. Crit. Care
Med. 164: 847–851.
29. Wu, A. K. L., V. C. C. Cheng, B. S. F. Tang, I. F. N. Hung, R. A. Lee, D. S. Hui,
and K. Y. Yuen. 2004. The unmasking of Pneumocystis jiroveci pneumonia dur-
ing reversal of immunosuppression: case reports and literature review. BMC In-
fect. Dis. 4: 57.
30. Cheng, V. C. C., I. F. N. Hung, A. K. L. Wu, B. S. F. Tang, C. M. Chu, and
K. Y. Yuen. 2004. Lymphocyte surge as a marker for immunorestitution disease
due to Pneumocystis jiroveci pneumonia in HIV-negative immunosuppressed
hosts. Eur. J. Clin. Microbiol. Infect. Dis. 23: 512–514.
31. Beck, J. M., M. J. Rosen, and H. H. Peavy. 2001. Pulmonary complications of
HIV infection: Report of the Fourth NHLBI Workshop. Am. J. Respir. Crit. Care
Med. 164: 2120–2126.
32. Wright, T. W., C. J. Johnston, A. G. Harmsen, and J. N. Finkelstein. 1999.
Chemokine gene expression during Pneumocystis carinii driven pulmonary in-
flammation. Infect. Immun. 67: 3452–3460.
33. Wright, T. W., F. Gigliotti, J. N. Finkelstein, J. T. McBride, C. L. An, and
A. G. Harmsen. 1999. Immune-mediated inflammation directly impairs pulmo-
nary function, contributing to the pathogenesis of Pneumocystis carinii pneumo-
nia. J. Clin. Invest. 104: 1307–1317.
34. Bhagwat, S. P., F. Gigliotti, H. D. Xu, and T. W. Wright. 2006. Contribution of
T cell subsets to the pathophysiology of Pneumocystis-related immunorestitution
disease. Am. J. Physiol. 291: L1256–L1266.
35. Chen, W. X., J. W. Mills, and A. G. Harmsen. 1992. Development and resolution
of Pneumocystis-carinii pneumonia in severe combined immunodeficient mice:
morphological study of host inflammatory responses. Int. J. Exp. Pathol. 73:
36. Roths, J. B., and C. L. Sidman. 1992. Both immunity and hyperresponsiveness to
Pneumocystis carinii result from transfer of CD4?but not CD8?T-cells into
severe combined immunodeficiency mice. J. Clin. Invest. 90: 673–678.
37. Shellito, J. E., V. V. Suzara, W. Blumenfeld, J. M. Beck, H. J. Steger, and
T. H. Ermak. 1990. A new model of Pneumocystis carinii infection in mice
selectivity depleted of helper T lymphocytes. J. Clin. Invest. 85: 1686–1693.
38. Beck, J. M., and A. G. Harmsen. 1998. Lymphocytes in host defense against
Pneumocystis carinii. Semin. Respir. Infect. 13: 330–338.
39. Harmsen, A. G., and M. Stankiewicz. 1990. Requirement for CD4?cells in
resistance to Pneumocystis carinii pneumonia in mice. J. Exp. Med. 172:
40. Beers, M. F., S. R. Bates, and A. B. Fisher. 1992. Differential extraction for the
rapid purification of bovine surfactant protein B. Am. J. Physiol. 262:
41. Beers, M. F., A. Wali, M. F. Eckenhoff, S. I. Feinstein, J. H. Fisher, and
A. B. Fisher. 1992. An antibody with specificity for surfactant protein C precur-
sors: identification of pro-SP-C in rat lung. Am. J. Respir. Cell Mol. Biol. 7:
42. Beck, J. M., R. L. Newbury, B. E. Palmer, M. L. Warnock, P. K. Byrd, and
H. B. Kaltrieder. 1996. Role of CD8?lymphocytes in host defense against Pneu-
mocystis carinii in mice. J. Lab. Clin. Med. 128: 477–487.
43. Beck, J. M., H. D. Liggit, E. N. Brunette, H. J. Fuchs, J. E. Shellito, and
R. J. Debs. 1991. Reduction in intensity of Pneumocystis carinii pneumonia in
mice by aerosol administration of gamma interferon. Infect. Immun. 59:
44. Rudmann, D. G., A. M. Preston, M. W. Moore, and J. M. Beck. 1998. Suscep-
tibility to Pneumocystis carinii in mice is dependent on simultaneous deletion of
IFN-? and type 1 and 2 TNF receptor genes. J. Immunol. 161: 360–366.
45. Zheng, M. Q., J. E. Shellito, L. Marrero, Q. Zhong, S. Julian, P. Ye, V. Wallace,
P. Schwarzenberger, and J. K. Kolls. 2001. CD4?T cell-independent vaccination
against Pneumocystis carinii in mice. J. Clin. Invest. 108: 1469–1474.
46. Phares, T. W., R. B. Kean, T. Mikheeva, and D. C. Hooper. 2006. Regional
differences in blood-brain barrier permeability changes and inflammation in the
apathogenic clearance of virus from the central nervous system. J. Immunol. 176:
47. Jassar, A. S., E. Suzuki, V. Kapoor, J. Sun, M. B. Silverberg, L. M. Cheung,
M. D. Burdick, R. M. Strieter, L. M. Ching, L. R. Kaiser, and S. M. Albelda.
2005. Activation of tumor-associated macrophages by the vascular disrupting
agent 5,6-dimethylxanthenone-4-acetic acid induces an effective CD8?T-cell-
mediated antitumor immune response in murine models of lung cancer and me-
sothelioma. Cancer Res. 65: 11752–11761.
48. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of
microgram quantities of protein utilizing the principle of protein-dye binding.
Ann. Biochem. 72: 248–254.
2286IMMUNE RECONSTITUTION AND SURFACTANT IN PCP
49. Bartlett, G. R. 1959. Phosphorous assay in column chromatography. J. Biol. Download full-text
Chem. 234: 466–468.
50. Atochina, E. N., A. J. Gow, J. M. Beck, A. Haczku, A. Inch, H. Kadire, Y. Tomer,
C. Davis, A. M. Preston, F. Poulain, et al. 2004. Surfactant protein D deficient
mice exhibit delayed clearance of Pneumocystis lung infection with increased
inflammation and altered nitric oxide metabolism. J. Infect. Dis. 189: 1528–1539.
51. Enhorning, G. 2001. Pulmonary surfactant function studied with the pulsating
bubble surfactometer (PBS) and the capillary surfactometer (CS). Comp. Bio-
chem. Physiol. A Mol. Integr. Physiol. 129: 221–226.
52. Jaffrey, S. R., and S. H. Snyder. 2001. The biotin switch method for the detection
of S-nitrosylated proteins. Sci. STKE 2001: pl1.
53. Casey, J., J. Kaplan, E. N. Atochina-Vasserman, A. J. Gow, H. Kadire, Y. Tomer,
J. H. Fisher, S. Hawgood, R. C. Savani, and M. F. Beers. 2005. Alveolar sur-
Am. J. Respir. Crit. Care Med. 172: 869–877.
54. Fang, K. Z., N. V. Ragsdale, R. M. Carey, T. Macdonald, and B. Gaston. 1998.
Reductive assays for S-nitrosothiols: implications for measurements in biological
systems. Biochem. Biophys. Res. Commun. 252: 535–540.
55. Meissner, N. N., S. Swain, M. Tighe, A. Harmsen, and A. Harmsen. 2005. Role
of type I IFNs in pulmonary complications of Pneumocystis murina infection.
J. Immunol. 174: 5462–5471.
56. Shellito, J. E., C. Tate, S. Ruan, and J. Kolls. 2000. Murine CD4?T lymphocyte
subsets and host defense against Pneumocystis carinii. J. Infect. Dis. 181:
57. Pottratz, S., and A. L. Weir. 1995. Attachment of Pneumocystis carinii to primary
cultures of rat alveolar epithelial cells. Exp. Cell Res. 221: 357–362.
58. Beck, J. M., A. M. Preston, J. G. Wagner, S. E. Wilcoxen, P. Hossler,
S. R. Meshnick, and R. Paine. 1998. Interaction of rat Pneumocystis carinii and
rat alveolar epithelial cells in vitro. Am. J. Physiol. 19: L118–L125.
59. Garvy, B. A., J. A. Wiley, F. Gigliotti, and A. G. Harmsen. 1997. Protection
against Pneumocystis carinii pneumonia by antibodies generated from either T
helper 1 or T helper 2 responses. Infect. Immun. 65: 5052–5056.
60. Shellito, J. E., J. K. Kolls, R. Olariu, and J. M. Beck. 1996. Nitric oxide and host
defense against Pneumocystis carinii infection in a mouse model. J. Infect. Dis.
61. Ischiropoulos, H., M. F. Beers, S. T. Ohnishi, D. Fisher, S. Garner, and
S. R. Thom. 1996. Nitric oxide production and perivascular tyrosine nitration in
brain following carbon monoxide poisoning in the rat. J. Clin. Invest. 97:
62. Genovese, T., S. Cuzzocrea, R. Di Paola, M. Failla, E. Mazzon, M. Sortino,
G. Frasca, E. Gili, N. Crimi, A. Caputi, and C. Vancheri. 2005. Inhibition or
knock out of inducible nitric oxide synthase result in resistance to bleomycin-
induced lung injury. Respir. Res. 6: 58.
63. Borron, P., R. A. Veldhuizen, J. F. Lewis, F. Possmayer, A. Caveney, K. Inchley,
R. G. McFadden, and L. J. Fraher. 1996. Surfactant associated protein-A inhibits
human lymphocyte proliferation and IL-2 production. Am. J. Respir. Cell Mol.
Biol. 15: 115–121.
64. Borron, P., J. C. McIntosh, T. R. Korfhagen, J. A. Whitsett, J. Taylor, and
J. R. Wright. 2000. Surfactant-associated protein A inhibits LPS-induced cyto-
kine and nitric oxide production in vivo. Am. J. Physiol. 278: L840–L847.
65. Koptides, M., T. M. Umstead, J. Floros, and D. S. Phelps. 1997. Surfactant
protein A activates NF-kappa B in the THP-1 monocytic cell line. Am. J. Physiol.
66. Atochina, E. N., M. F. Beers, S. Hawgood, F. Poulain, C. Davis, T. T. Fusaro, and
A. J. Gow. 2003. Surfactant protein-D, a mediator of innate lung immunity, alters
the products of nitric oxide metabolism. Am. J. Respir. Cell Mol. Biol. 30:
67. Atochina, E. N., J. M. Beck, A. M. Preston, A. Haczku, Y. Tomer, S. T. Scanlon,
T. Fusaro, J. Casey, S. Hawgood, A. J. Gow, and M. F. Beers. 2004. Enhanced
lung injury and delayed clearance of Pneumocystis carinii in surfactant protein
A-deficient mice: attenuation of cytokine responses and reactive oxygen-nitrogen
species. Infect. Immun. 72: 6002–6011.
68. Hawgood, S., C. Brown, J. Edmondson, A. Stumbaugh, L. Allen, J. Goerke,
H. Clark, and F. Poulain. 2004. Pulmonary collectins modulate strain-specific
influenza A virus infection and host responses. J. Virol. 78: 8565–8572.
69. Clark, H., N. Palaniyar, P. Strong, J. Edmondson, S. Hawgood, and
K. B. M. Reid. 2002. Surfactant protein D reduces alveolar macrophage apoptosis
in vivo. J. Immunol. 169: 2892–2899.
70. Gardai, S. J., Y. Q. Xiao, M. Dickinson, J. A. Nick, D. R. Voelker, K. E. Greene,
and P. M. Henson. 2003. By binding SIRP? or calreticulin/CD91, lung collectins
act as dual function surveillance molecules to suppress or enhance inflammation.
Cell 115: 13–23.
71. Limper, A. H., D. M. O’Riordan, Z. Vuk-Pavlovic, and E. C. Crouch. 1994.
Accumulation of surfactant protein D in the lung during Pneumocystis carinii
pneumonia. J. Eukaryot. Microbiol. 41: 98S.
72. Beers, M. F., R. G. Johnson, and A. Scarpa. 1986. Evidence for an ascorbate
shuttle for the transfer of reducing equivalents across chromaffin granule mem-
branes. J. Biol. Chem. 261: 2529–2535.
73. Wang, J., F. Gigliotti, S. P. Bhagwat, S. B. Maggirwar, and T. W. Wright. 2007.
Pneumocystis stimulates MCP-1 production by alveolar epithelial cells through a
JNK-dependent mechanism. Am. J. Physiol. 292: L1495–L1505.
2287The Journal of Immunology