Characterization of extended co-culture of non-typeable Haemophilus influenzae with primary human respiratory tissues.

Abstract

Non-typeable Haemophilus influenzae (NTHi) are human-adapted Gram-negative bacteria that comprise part of the normal flora of the human upper airway, but are also responsible for a number of mucosal infections such as otitis media and bronchitis. These infections often recur and can become chronic. To characterize the effect of long-term co-culture of NTHi with human tissues, we infected primary respiratory epithelial cells grown at the air-liquid interface with three NTHi strains over a range of 1-10 days. Scanning and transmission electron microscopy of tissues confirmed that intact NTHi were persisting paracellularly, while organisms observed in intracellular vacuoles appeared degraded. Furthermore, the apical surface and tight junctions of the infected tissues were undisturbed, with high transepithelial electrical resistances, while the basal cell layer displayed more junctional disorganization and wider intercellular spaces than the uninfected control tissues. Although the tissues elaborated the cytokine profile reported for NTHi-caused otitis media in vivo, there was little change in the dynamics of cytokine secretion over the time points tested. Finally, we report that NTHi strains released outer membrane vesicles (OMVs) during extended co-culture with the tissues, and show that these OMVs directly interact with host cell membranes.

Full text

Page 1

Original Research
Characterization of extended co-culture of non-typeable
Haemophilus influenzae with primary human respiratory tissues
Dabin Ren1, Kevin L Nelson2, Peter N Uchakin1, Arnold L Smith2, Xin-Xing Gu3and
Dayle A Daines1
1Division of Basic Medical Sciences, Mercer University School of Medicine, 1550 College Street, Macon, GA 31207;2Seattle Children’s
Research Institute, 1900 Ninth Avenue, C9S-8, Seattle, WA 98101;3National Institute on Deafness and Other Communication Disorders,
5 Research Court, Rockville, MD 20850, USA
Corresponding author: Dayle A Daines. Email: daines_da@mercer.edu
Abstract
Non-typeable Haemophilus influenzae (NTHi) are human-adapted Gram-negative bacteria that comprise part of the normal
flora of the human upper airway, but are also responsible for a number of mucosal infections such as otitis media and
bronchitis. These infections often recur and can become chronic. To characterize the effect of long-term co-culture of
NTHi with human tissues, we infected primary respiratory epithelial cells grown at the air–liquid interface with three NTHi
strains over a range of 1–10 days. Scanning and transmission electron microscopy of tissues confirmed that intact NTHi
were persisting paracellularly, while organisms observed in intracellular vacuoles appeared degraded. Furthermore, the
apical surface and tight junctions of the infected tissues were undisturbed, with high transepithelial electrical resistances,
while the basal cell layer displayed more junctional disorganization and wider intercellular spaces than the uninfected
control tissues. Although the tissues elaborated the cytokine profile reported for NTHi-caused otitis media in vivo, there
was little change in the dynamics of cytokine secretion over the time points tested. Finally, we report that NTHi strains
released outer membrane vesicles (OMVs) during extended co-culture with the tissues, and show that these OMVs directly
interact with host cell membranes.
Keywords: NTHi, NHBE, co-culture, persistence
Experimental Biology and Medicine 2012; 237: 540–547. DOI: 10.1258/ebm.2012.011377
Introduction
Non-typeable Haemophilus influenzae (NTHi) are small fasti-
dious Gram-negative organisms that have evolved to be obli-
gate parasites that constitute part of the normal flora of the
human upper airway. In addition to being commensals,
these bacteria are implicated in mucosal infections, including
otitis media and bronchitis.1,2Such infections can be recurrent
and resistant toantimicrobial therapy, often requiring sequen-
tial treatments and, in the case of chronic otitis media with
effusion, surgical intervention. Lower respiratory tract infec-
tions such as pneumonia and exacerbations of chronic
obstructive pulmonary disease have also been attributed to
NTHi.3,4Furthermore, there is evidence that NTHi can form
biofilms on mucosal surfaces in vivo.5Taken together, these
characteristics suggest that this organism can adapt to a
number of different lifestyles, depending upon its immediate
microenvironment in the host as well as other factors.
The traditional method of investigating how a pathogen
interacts with human cells has been to grow epithelial
cells as a submerged monolayer in vitro and then challenge
with the pathogen. Various specific attributes, such as bac-
terial adherence and cell entry, can be measured by harvest-
ing and enumerating either the total cell-associated bacteria
(those that were adherent and those that had been interna-
lized) or only those that had survived after the monolayer
had been treated with gentamicin.6Since this antibiotic
does not enter human cells during the course of the assay,
only those bacteria that were inside the human cells prior
to the gentamicin treatment can survive. Using this
approach, early events in the interaction of NTHi with
human cells can be efficiently and effectively quantified.
However, since NTHi can also grow in the overlying cell
culture media as well as enter the epithelial cells, this
assay is limited in the amount of time that the monolayer
can be co-cultured with the bacteria (usually hours) before
unacceptable human cell toxicity results. In addition, epi-
thelial cells do not differentiate or polarize well in a
submerged culture.7These issues were addressed by a sub-
sequent model that consists of first growing primary epi-
thelial cells on a porous membrane of a tissue culture
ISSN: 1535-3702
Copyright # 2012 by the Society for Experimental Biology and Medicine
Experimental Biology and Medicine 2012; 237: 540–547

Page 2

insert, with media above and below the membrane. After
the cells reach confluence, the apical media is removed.
This raises the epithelial cells to the air–liquid interface
(ALI) and allows the cells to be fed by media from the
basal side.8In this study, primary human respiratory epi-
thelial cells at the ALI are allowed to differentiate and polar-
ize for a number of days before being co-cultured with
NTHi, which are inoculated directly onto the apical
surface. Particularly when using primary human respiratory
epithelial cells, this approach is both quite suitable for long-
term co-culture with NTHi and is biologically relevant, as it
is an excellent in vitro representation of human upper
airway tissue.7In addition, since the epithelium of the
middle ear is part of the respiratory mucosa and produces,
as do bronchial epithelial cells, a complex mixture of mucin
and cytokines, this model is appropriate for the study of
otitis media pathogens.9We hypothesized that NTHi
causes an acute inflammatory response upon contact, and
investigated the effect of co-culture on the tissues. Herein,
we present data on how and where NTHi survives over
time within this ALI model, how the tissues react to long-
term bacterial infection and the maintenance of transepithe-
lial electrical resistance (TER) following NTHi co-culture
with both normal human bronchial epithelial (NHBE) cells
(Lonza, Walkersville, MD, USA) at the ALI, and a commer-
cially available tissue model, the EpiAirwayTM(MatTek,
Ashland, MA, USA). The EpiAirway tissues consist of
primary, well-differentiated,
with tight junctions, ciliated and non-ciliated cells, goblet
cells that produce mucin, and which retain the ability to
produce cytokines in response to infection.
three-dimensional tissues
Materials and methods
Bacterial cultures
All bacterial strains used in this study are pertinent to otitis
media. Strain R2846 was isolated from the middle ear of a
child with otitis media.10Strain R2866 was isolated from
the blood of a child with otitis media11and strain
86-028NP was isolated from the nasopharynx of a child
with otitis media.12The genome of each of the strains has
been sequenced and annotated, facilitating future compara-
tive studies. NTHi strains were grown overnight at 378C in
5% CO2on chocolate agar plates with 5 U/mL bacitracin, or
brain heart infusion agar supplemented with 10 mg/mL of
b-NAD and 10 mg/mL of hemin chloride (sBHI). Broth cul-
tures were grown in sBHI at 378C with shaking in room air.
ALI NHBE cell cultures
For NHBE cell (#CC-2540; Lonza) ALI cultures, 5.0–7.0 ? 105
NHBE cells were seeded and grown submerged on BD
Biocoat membranes with a 4.2 cm2growth area and a 3 mm
pore size in 50/50 bronchial epithelial cell basal medium
(#CC-3171; Lonza)/Dulbecco’s modified Eagle’s medium
(#10 013 CM; Mediatech cellgro, Manassas, VA, USA) with
bronchial epithelial growth medium SingleQuots (#CC-4175;
Lonza) containing retinoic acid (#R2625; Sigma, St Louis,
MO, USA) at 10? the reference concentration13until
confluence was reached. This medium is similar to Matsui-2
described by Sachs et al.14and is hereinafter referred to as
TC medium. At this time, the apical medium was removed
and 2.5 mL of TC medium was added to the basal compart-
ment. The TER was measured using EVOM2 (World
Precision Instruments, Sarasota, FL, USA) according to the
manufacturer’s instructions. Between three and five days
afterbeing placed at ALI,theTERoftheNHBEcellsstabilized
at ?400 V cm2, which starts to decrease after several weeks at
the ALI. The apical side of each insert was subsequently
inoculated with ?1.0 ? 103colony-forming units (CFU) of
NTHi strain R2846 in a volume of 200 mL in Dulbecco’s
phosphate-buffered saline (D-PBS). Data are representative
of a total of six different lots of NHBE cells.
A total of seven different lots of EpiAirway tissues
(#AIR-100-ABF or #AIR-606-ABF; MatTek, Ashland, MA,
USA) were grown on collagen-coated Millipore MillicellTM
CM (Millipore, Billerica, MA, USA) single-well tissue
culture plate inserts with a pore size of 0.4 mm and received
22 days after seeding. After at least 24 h of equilibration at
378C in 5% CO2, each insert was inoculated with NTHi
strains R2846, R2866 or 86-028NP. The initial TER of each
tissue was ?250 V cm2. The proprietary basal medium
(#AIR-100-MM; MatTek) was changed and the apical
surfaces were washed with 200 mL of D-PBS on a daily basis.
ALI co-cultures
Overnight cultures of NTHi strains were resuspended from
chocolate or sBHI agar plates into D-PBS at an optical
density (600 nm) of ?0.2. These initial solutions were
diluted in D-PBS to the desired concentrations (?1.0 ?
103–1.0 ? 107CFU) and used to apically inoculate ALI
NHBE cells or EpiAirway tissues. The 0.6 cm2growth area
inserts were inoculated with ?25 mL, while the 4.2 cm2
inserts received 200 mL of the bacterial suspensions.
ALI harvest
At the desired time points, each insert was washed 3? with
D-PBS, and incubated both apically and basally with TC or
basal media containing 100 mg/mL gentamicin for one hour
at 378C. The inserts were washed 3? with D-PBS without
calcium and magnesium, and then incubated for 10 min at
378C with 1% saponin in D-PBS without calcium and mag-
nesium. Following this, each insert was scraped and the
tissues were mechanically disintegrated, diluted and plated
for viable bacteria as described previously.15The surviving
organisms were expressed as gentamicin-resistant CFU/cm2
of the growth area.
Enzyme-linked immunosorbent assay assays
EpiAirwaywasheswerefirstanalyzedwiththeMulti-Analyte
ELISArray kits (SABioscienses, QIAGEN Inc, Valencia, CA,
USA) for exploratory detection of cytokines. Results were
read on a Tecan InfinitewM1000 Plate Reader (Tecan Group
Ltd, Ma ¨nnedorf, Switzerland) at 450 nm with l correction of
570 nm followed by acquisition and reduction with data
analysis software Magellan (Tecan Group Ltd) for consequent
................................................................................................................................................
Ren et al. Long-term co-culture with NTHi
541

Page 3

statistical analysis. Subsequently, electrochemiluminescence
detection in multi-plex format (Meso Scale Discovery,
Gaithersburg,MA,USA)wasusedforcytokinequantification.
Results were read on a MSD Sector Imager 2400E followed
by acquisition and reduction with data analysis software
Discovery Workbench 3.0.18 (Meso Scale Discovery).
Statistical analysis
All graphical data are expressed as the mean+SD of the
indicated number of experiments. The level of significance
was set at P, 0.05. The statistical software package
SigmaStat (Systat Software Inc, Chicago, IL, USA) was
used for statistical analysis of the collected enzyme-linked
immunosorbentassay (ELISA)
one-way analysis of variance on Ranks was implemented
to determine statistical differences between studied intervals
of sample collection for each cytokine.
data.Kruskal–Wallis
Immunoelectron microscopy
EpiAirway inserts were first co-cultured as described above
with NTHi strains R2866 or 86-028NP for 5 days prior to fix-
ation, or with D-PBS sham inoculation. The inserts were then
washed and immersed in 4% paraformaldehyde and 0.2% glu-
taraldehyde in 100 mmol/L sodium cacodylate buffer (pH 7.4)
foronehouratroomtemperature,followedby18 hat48C.Each
membrane was then removed from the plastic insert using a
#11 pointed scalpel blade, immersed in fixative and processed
by the Georgia Health Sciences University (GHSU) Electron
Microscopy Core (Augusta, GA, USA). After incubation with
anti-NTHiprimaryantibody followed by gold-labeled second-
aryantibody, the grids were stained briefly with uranyl acetate
and imaged using a JEOL JEM 1230 transmission electron
microscope (TEM; JEOL USA Inc, Peabody, MA, USA).
Transmission and scanning electron microscopy
Images of infected and uninfected (control) EpiAirway inserts
wereobtainedafterfixationin1.25%glutaraldehyde,2.0%par-
aformaldehyde in 100 mmol/L sodium cacodylate buffer (pH
7.2) for 24 h. The membranes were then removed from the
plastic insert using a #11 pointed scalpel blade, immersed in
fixative and processed by the GHSU Electron Microscopy
Core. Briefly, the fixed membranes were osmicated16and cut
in half. The TEM half was dehydrated, infiltrated with LR
White, sectioned and imaged using a JEOL JEM 1230 trans-
missionelectronmicroscope.Thescanningelectronmicroscopy
(SEM) half was critical-point dried with CO2, adhered to stubs
andsputter-coatedwithgold–palladiumpriortoimagingona
FEI XL30 field emission gun SEM (FEI Company, Hillsboro,
OR, USA). This resulted in paired SEM and TEM images of
the same insert.
Results
NTHi strains survive within ALI tissues for extended
periods of time
To determine the length of time that NTHi could be
co-cultured with ALI tissues, we tested the ability of
various NTHi strains to survive within ALI NHBE cells
and EpiAirway tissues over time points ranging from 1 to
10 days (Figure 1). At each desired time point, the tissues
were harvested and the gentamicin-resistant (internalized)
bacteria were enumerated by plate counts. Figure 1a
showsthe numbersof
EpiAirway tissues from long-term co-cultures, expressed
as viable gentamicin-resistant CFU recovered per square
centimeter of tissue (GmRCFU/cm2), for NTHi strain
86-028NP. These data are consistent with previous results
observed for strain R2866.15Figure 1b illustrates the GmR
bacteria surviving within
Figure 1
NHBE cells. (a) Internalized (gentamicin-resistant) bacteria of strain 86-028NP
recovered fromEpiAirwaytissues,
(b) Internalized strain R2846 recovered from ALI NHBE cells, 1–5 days after
infection (n ¼ 2). (c) Internalized strain R2846 recovered from EpiAirway
tissues, 1–8 days after infection (n ¼ 2). Error bars are SD. ALI, air–liquid
interface; NHBE, normal human bronchial epithelial; NTHi, non-typeable
Haemophilus influenzae
Survival of NTHi strains over time inside EpiAirway tissues and ALI
2–10 days afterinfection (n ¼ 6).
................................................................................................................................................
542
Experimental Biology and MedicineVolume 237May 2012

Page 4

CFU/cm2of strain R2846 in ALI cultures of NHBE cells, and
Figure 1c shows R2846 recovered from EpiAirway tissues.
This confirms that ALI cultures can be used to investigate
the long-term interactions of various strains of NTHi with
primary human epithelial cells in vitro.
Long-term co-culture of NTHi with primary tissues at
the ALI does not result in significant damage to the
apical cell layer
To examine the effect of extended NTHi co-culture on
EpiAirway tissues, SEM and TEM images were taken of
inserts with and without infection (Figure 2). These SEM
and TEM images are paired, as each EpiAirway tissue was
divided, with one half processed for SEM and the other
half processed for TEM. Figure 2a shows a representative
SEM image of tissue infected with strain R2866 for five
days, while Figure 2b shows a SEM of the uninfected
control EpiAirway tissue. There was no apparent difference
in the number of ciliated cells or the length and number of
microvilli between theinfected
However, there appeared to be more goblet cells in the
infected tissues than in the control, with 70 goblet cells
observed in 10 random SEM images of infected tissues,
versus 46 identified in the same number of random
images of uninfected tissues (each at ?5000 magnification).
To image any internal effects of NTHi co-culture on
EpiAirway tissues, TEM was performed on each sample.
Figure 2c shows a representative TEM image of tissue
and controltissues.
infected for five days with strain R2866, and Figure 2d
shows the uninfected control tissue. Consistent with the
SEM results, cells at the apical surface of the infected
tissues were undamaged, with tight junctions and adherens
junctions that appeared intact. However, the basal cell layer
of the infected tissue exhibited more disorganized junctions,
with wider intercellular spaces, than those in the control
tissue.
Infected tissues maintain polarization during
extended periods of co-culture
After infection, the TER, a measure of functional tight junc-
tions, was quantified over time in both EpiAirway tissues
(Figure 3a) and ALI NHBE cells (Figure 3b) infected with
strain R2846. Average TER values expressed in V cm2are
reported. In both models, no significant decrease in TER
was observed over the course of the co-culture with NTHi.
In a separate experiment, the TER of two ALI NHBE cell cul-
tures was measured prior to infection with strain R2846, and
again eight days after inoculation. To disrupt the TER in
these infected ALI NHBE cell cultures, 0.2 mL of 0.02% ethyl-
ene glycol tetraacetic acid was added to the apical surface of
each insert for four hours. The TERs decreased from more
than 400 to 20 and 22 V cm2, respectively, indicating that
the measured electrical resistance was due to functional
tight junctions. These data support the SEM and TEM find-
ings and confirm that extended co-culture with NTHi does
not disturb epithelial cell tight junction integrity.
Figure 2
One EpiAirway insert was infected with strain R2866 for five days, then fixed and cut in half. One half was processed for SEM (a), the other for TEM (c). Another
EpiAirway insert was not infected with NTHi and used as the control for SEM (b) and TEM (d). (a) SEM of infected EpiAirway tissue. (b) SEM of control EpiAirway
tissue. (c) TEM of infected EpiAirway tissue. (d) TEM of control EpiAirway tissue. NTHi, non-typeable Haemophilus influenzae
Paired scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of EpiAirway tissues with or without NTHi co-culture.
................................................................................................................................................
Ren et al. Long-term co-culture with NTHi
543

Page 5

Infected tissues elaborate proinflammatory cytokines
that match those reported during in vivo infections
To determine whether the ALI tissues expressed the epi-
thelial cell cytokine profile during long-term infections
with NTHi in vitro that has been reported during infections
in vivo,17ELISA assays were performed on the pooled apical
washes of EpiAirway tissues co-cultured with strain
86-028NP at day 0 (prior to infection) and at two, four
and six days after infection (Figure 4). Interleukin-8 (IL-8)
was the predominant cytokine, followed by IL-6, IL-1
andgranulocyte/macrophage
(GM-CSF).Uninfected EpiAirway
approximately 200 ng/mL of IL-8 (Figure 4, day 0 time
point), as airway epithelium produced IL-8 on a constitutive
basis.18
No significant amounts of IL-2, IL-4, IL-10,
IL-12p40, IL-17 or interferon-g were detected in the
samples. Further, there were no statistically significant
changes in the dynamics of the secretion of tumor necrosis
factor-a, IL-1b, IL-6, IL-8, and GM-CSF in the apical
washes over the time points tested. This is consistent with
a previous study showing that the mRNA levels of
proinflammatory cytokines elicited by NTHi infections
in vivo peaked at 3–6 hours after infection, and prior to
the first clinical signs of otitis media.19
In a separate set of experiments using NHBE cells at the
ALI, we assayed both the apical and basal compartments
for the presence of IL-8 by ELISA at one day and eight
days after inoculation with strain R2846. At one day, the
average concentration of IL-8 in the apical washes was
251 ng/mL, while the concentration in the basal media was
colony-stimulating
tissues
factor
produced
69 ng/mL. At eight days, the average IL-8 concentration was
350 ng/mL apically and 120 ng/mL in the basal media.
Thus, the basal compartment concentration of IL-8 was
approximately 28% to34% of the apical values in thissystem.
NTHi long-term co-culture with EpiAirway tissues
results in focal areas of infection
Immunoelectron microscopy (IEM) with anti-NTHi anti-
serum followed by gold-labeled secondary antibodies of
EpiAirway tissues after five days of infection with strains
R2866 and 86-028NP demonstrated that NTHi were located
in focal sites within the EpiAirway tissues, rather than exist-
ing as a uniform infection throughout the tissues (approxi-
mately 80 organisms are located between the cells shown
in Figure 5a). The organisms observed within intracellular
membrane-bound vacuoles appeared to be degraded, and
IEMconfirmedthatthesevacuolescontainedNTHi (approxi-
mately 21 vacuoles reacted with the anti-NTHi antiserum
shown in Figure 5b). The majority of the intact organisms
visualized were located between the epithelial cells in the
basal cell layers (Figure 5c). The paracellular location of
NTHifollowinginfectionofimmortalizedhumanrespiratory
epithelial cells has been observed previously,20and the pres-
ence of NTHi within intracellular vacuoles in primary cul-
tures has also been reported.21However, the duration of the
co-culture of NTHi with the human cells for both of these
studies was 30 h or less. Our data reveal that, over extended
periods of co-culture with these ALI tissues, the location of
intactNTHi isprimarilyparacellular,with intracellularvacu-
oles containing organisms in various stages of degradation.
NTHi release outer membrane vesicles within the
EpiAirway tissues
We observed that both strains R2866 and 86-028NP released
outermembranevesicles (OMVs) during long-term
Figure 3
NHBE (b) cell cultures. (a) EpiAirway cultures were inoculated with strain
R2846 and the TER was measured during co-culture. (b) NHBE cell cultures
at the ALI were inoculated with strain R2846 and the TER was measured
during co-culture. Horizontal axes denote time after inoculation for each
reading. Error bars are SD. ALI, air–liquid interface; NHBE, normal human
bronchial epithelial
Transepithelial electrical resistance (TER) of EpiAirway (a) and ALI
Figure 4
over time. ELISA assays were performed to detect the levels of IL-1b, IL-6,
TNF-a, GM-CSF and IL-8 in pooled EpiAirway washes (n ¼ 2) during co-culture.
Day 0 indicates assays prior to inoculation, and the following days indicate
length of time of co-culture. IL, interleukin; GM-CSF, granulocyte/macrophage
colony-stimulating factor; ELISA, enzyme-linked immunosorbent assay; TNF,
tumor necrosis factor
Cytokine profile of EpiAirway inserts infected with strain 86-028NP
................................................................................................................................................
544
Experimental Biology and Medicine Volume 237May 2012

Page 6

co-culture with EpiAirway tissues (Figure 6). The vesicles
appeared to be shed when the organisms were paracellular,
with some vesicles associating with the lateral membrane of
the epithelial cells (Figure 6a). These OMVs were labeled
with anti-NTHi antiserum and were identified by IEM, indi-
cating that they originated from the NTHi strains (Figure 5c,
arrowhead; Figure 6b, arrows). The role of these OMVs
during long-term infection is currently under investigation.
Discussion
The normal lifestyle of NTHi is as a human commensal, and
its role as a pathogen can be considered to be accidental or
opportunistic. Due to these unique characteristics, investi-
gators are interested in determining how NTHi interact
with respiratory epithelial cells, the site of colonization
by H. influenzae, which can lead to superficial or invasive
infection as a result of this interaction. Many important
discoveries regarding NTHi pathogenic mechanisms and
interactions with host cells have been made using sub-
merged monolayer models,22–24but their utility for long-
term co-culture is arguably less than using a differentiated
tissue model at the ALI. Understanding how NTHi
survive during chronic mucosal infections, particularly
otitis media and bronchitis, necessitates the ability to
observe the organisms interacting with and persisting
within host tissues over extended periods of time.
Although an immortalized cell line can be used to form
the basis of the ALI model, primary human respiratory epi-
thelial cells are particularly well suited for the study of
NTHi infections of days or weeks duration. The detection
of the major proinflammatory cytokines of the innate
immune response during co-culture with NTHi suggests
the adequate immune competence of the cells composing
the tissues, even though many elements of the innate
immune response and all of the adaptive are not present.
This model allows investigators to study the interaction of
NTHi within host tissues without the requirement to
obtain Institutional Review Board approval, as the investi-
gator is blinded to donor information by the company sup-
plying the tissues and/or primary cells. Primary cells can
also be pooled, decreasing the possibility of single-donor
genetic factors influencing a particular lot of tissues.
Finally, these ALI tissues are ideal for studying the
long-term host–pathogen interactions, particularly for the
in vitro modeling of recurrent diseases such as otitis media
and lung infections.
It is known based on clinical observations in which chil-
dren had nasopharyngeal colonization without signs or
Figure 5
86-028NP are paracellular (a, arrows). The bacteria that are intracellular appear to be degraded (boxed area). N ¼ nucleus. (b) Magnified view of the boxed
area in (a). Arrows indicate the intracellular and degraded bacteria, which react with the anti-NTHi antiserum. N ¼ nucleus. (c) Paracellular organisms of strain
R2866. Arrows indicate intact bacteria; arrowhead indicates an outer membrane vesicle (OMV) that reacts with anti-NTHi antiserum. NTHi, non-typeable
Haemophilus influenzae
Immunoelectron microscopy of EpiAirway tissues co-cultured for five days with strains 86-028NP or R2866. Following co-culture, most of the intact
................................................................................................................................................
Ren et al. Long-term co-culture with NTHi
545

Page 7

symptoms of inflammation that NTHi strains can interact
with host cells for months without eliciting an appreciable
inflammatory response,25and these organisms have been
observedgrowingasbiofilms
in vivo.5However, it has also been well documented that
shortly after inoculation, NTHi strains adhere to, and
invade, host cells in vitro.6,20,21,26This paradox suggests
that there are mechanisms by which NTHi can survive
within the tissues over time without overtly damaging the
epithelium and triggering host defenses. Our data, although
limited to days rather than months, indicate that NTHi
strains form focal communities within tissues at the ALI
that do not disturb the structure and function of the apical
cell layer or the TER of the tissues, and result in low levels
of epithelial cell proinflammatory cytokine expression.
This lifestyle makes sense for an organism whose survival
depends upon its ability to scavenge essential nutrients
without initiating its own destruction by the host response.
We also report here that NTHi release OMVs while per-
sisting inside host tissues. Most Gram-negative bacteria
shed these vesicles, both constitutively during normal
growth, and to a greater extent under stressful environ-
mental conditions.27OMVs from various organisms have
been shown to contain lipopolysaccharide and proteins as
well as DNA,28,29and a recent study has identified the pro-
teome of OMVs shed from NTHi during stationary phase
growth in rich media in vitro.30A number of reasons for
OMV shedding have been proposed, including niche com-
petition with other bacteria, as a mechanism of horizontal
gene transfer, as a method of antigen delivery to host
cells, and as quorum-sensing molecule shuttles in bio-
films.31,32Our data demonstrate that NTHi shed these
OMVs during long-term infections within primary human
respiratory tissues at the ALI and confirm that the vesicles
can, and do, interact with host cell membranes in this model.
We conclude that the results of this study provide direct
and novel evidence of how NTHi interact with, and
survive within, host tissues over time. During long-term
co-culture with primary human respiratory tissues at the
ALI, NTHi strains do not elicit a significant inflammatory
on mucosalsurfaces
response, as measured by a substantial increase in cytokine
release, or damage the apical surfaces. Further, TER
measurements confirm that extended infection with NTHi
does not interfere with, or disrupt, junctional integrity.
Interestingly, we found that the organisms survived
within the tissues in paracellular foci of infection, with
those observed located in intracellular vacuoles appearing
to be degraded. Finally, we showed that NTHi release
OMVs during long-term infection in this model, and
that these vesicles can bind to host cell membranes.
Experiments are ongoing to determine the role of these
vesicles during NTHi infections.
Author contributions: All authors participated in the
design, interpretation of the data and review of the
manuscript. DR, KLN, PNU and DAD performed the exper-
iments; X-XG provided critical reagents (a-NTHi anti-
bodies); and DAD, PNU and ALS wrote the manuscript.
ACKNOWLEDGEMENTS
We thank Dr Robert S Munson Jr for strain 86-028NP, Dr
Robert J McKallip for critical reading of the manuscript,
and Robert M Smith and Libby Perry of the Georgia
Health Sciences University Electron Microscopy Core for
their excellent technical skills. This study was funded by
Public Health Service grants AI044002 from the National
Institute of Allergy and Infectious Diseases to ALS and
DC010187 from the National Institute on Deafness and
Other Communication Disorders to DAD.
REFERENCES
1 Murphy TF, Faden H, Bakaletz LO, Kyd JM, Forsgren A, Campos J, Virji
M, Pelton SI. Nontypeable Haemophilus influenzae as a pathogen in
children. Ped Infect Dis J 2009;28:43–8
2 Bandi V, Apicella MA, Mason E, Murphy TF, Siddiqi A, Atmar RL,
Greenberg SB. Nontypeable Haemophilus influenzae in the lower
respiratory tract of patients with chronic bronchitis. Am J Respir Crit Care
Med 2001;164:2114–9
Figure 6
tissues. The bacteria are paracellular and are releasing OMVs (arrows), some of which appear to interact with the host cell membrane. (b) IEM which shows
strain 86-028NP co-cultured with EpiAirway tissues for five days releasing OMVs that react with anti-NTHi antiserum (arrows) while located paracellularly.
NTHi, non-typeable Haemophilus influenzae; TEM, transmission electron microscopy; IEM, immunoelectron microscopy
NTHi release outer membrane vesicles while co-cultured with EpiAirway tissues. (a) TEM of strain R2866 after a five-day co-culture with EpiAirway
................................................................................................................................................
546
Experimental Biology and MedicineVolume 237 May 2012

Page 8

3 van Wessel K, Rodenburg GD, Veenhoven RH, Spanjaard L, van der
Ende A, Sanders EA. Nontypeable Haemophilus influenzae invasive
disease in the Netherlands: a retrospective surveillance study 2001–2008.
Clin Infect Dis 2011;53:e1–7
4 Chin CL, Manzel LJ, Lehman EE, Humlicek AL, Shi L, Starner TD,
Denning GM, Murphy TF, Sethi S, Look DC. Haemophilus influenzae from
patients with chronic obstructive pulmonary disease exacerbation induce
more inflammation than colonizers. Am J Respir Crit Care Med
2005;172:85–91
5 Hall-Stoodley L, Hu FZ, Gieseke A, Nistico L, Nguyen D, Hayes JD,
Forbes M, Greenberg DP, Dice B, Burrows A, Wackym PA, Stoodley P,
Post JC, Ehrlich GD, Kerschner JE. Direct detection of bacterial biofilms
on the middle-ear mucosa of children with chronic otitis media. JAMA
2006;296:202–11
6 St. Geme JW III. Molecular and cellular determinants of non-typeable
Haemophilus influenzae adherence and invasion. Cell Microbiol
2002;4:191–200
7 Pezzulo AA, Starner TD, Scheetz TE, Traver GL, Tilley AE, Harvey BG,
Crystal RG, McCray PG Jr, Zabner J. The air-liquid interface and use of
primary cell cultures are important to recapitulate the transcriptional
profile of in vivo airway epithelia. Am J Physiol Lung Cell Mol Physiol
2011;300:L25–L31
8 Whitcutt MJ, Adler KB, Wu R. A biphasic chamber system for
maintaining polarity of differentiation of cultured respiratory tract
epithelial cells. In vitro Cell Dev Biol 1988;24:420–8
9 Choi JY, Kim CH, Lee WS, Kim HN, Song KS, Yoon JH. Ciliary and
secretory differentiation of normal human middle ear epithelial cells.
Acta Otolaryngol 2002;122:270–5
10 Barenkamp SJ, Leininger E. Cloning, expression, and DNA sequence
analysis of genes encoding nontypeable Haemophilus influenzae
high-molecular-weight surface-exposed proteins related to filamentous
hemagglutinin of Bordetella pertussis. Infect Immun 1992;60:1302–13
11 Nizet V, Colina KF, Almquist JR, Reubens CE, Smith AL. A
virulent nonencapsulated Haemophilus influenzae. J Infect Dis
1996;173:180–6
12 Harrison A, Dyer DW, Gillaspy A, Ray WC, Mungur R, Carson MB,
Zhong H, Gipson J, Gipson M, Johnson LS, Lewis L, Bakaletz LO,
Munson RS Jr. Genomic sequence of an otitis media isolate of
nontypeable Haemophilus influenzae: comparative study with H. influenzae
serotype d, strain KW20. J Bacteriol 2005;187:4627–36
13 Linderholm AL, Onitsuka J, Xu C, Chiu M, Lee WM, Harper RW.
All-trans retinoic acid mediates DUOX2 expression and function in
respiratory tract epithelium. Am J Physiol Lung Cell Mol Physiol
2010;299:L215–21
14 Sachs LA, Finkbeiner WE, Widdicombe JH. Effects of media on
differentiation of cultured human tracheal epithelium. In Vitro Cell Dev
Biol Anim 2003;39:56–62
15 Ren D, Daines DA. Use of the EpiAirway model for characterizing
long-term host-pathogen interactions. J Vis Exp 2011;(55):e3261
16 Nanci A, Mazariegos M, Fortin M. The use of osmicated tissues for
Lowicryl K4M embedding. J Histochem Cytochem 1992;40:869–74
17 Tong HH, Chen Y, Liu X, DeMaria TF. Differential expression of cytokine
genes and iNOS induced by nonviable nontypeable Haemophilus
influenzae or its LOS mutants during acute otitis media in the rat. Int J
Pediatr Otorhinolaryngol 2008;72:1183–91
18 McDougall CM, Blaylock MG, Douglas JG, Brooker RJ, Helms PJ, Walsh
GM. Nasal epithelial cells as surrogates for bronchial epithelial cells in
airway inflammation studies. Am J Respir Cell Mol Biol 2008;39:560–8
19 Melhus A, Ryan AF. Expression of cytokine genes during pneumococcal
and nontypeable Haemophilus influenzae acute otitis media in the rat.
Infect Immun 2000;68:4024–31
20 van Schilfgaarde M, van Alphen L, Eijk P, Everts V, Dankert J.Paracytosis
of Haemophilus influenzae through cell layers of NCI-H292 lung epithelial
cells. Infect Immun 1995;63:4729–37
21 Ketterer MR, Shao JQ, Hornick DB, Buscher B, Bandi VK, Apicella MA.
Infection of primary human bronchial epithelial cells by Haemophilus
influenzae: macropinocytosis as a mechanism of airway epithelial cell
entry. Infect Immun 1999;67:4161–70
22 Ratner AJ, Aguilar JL, Shchepetov M, Lysenko ES, Weiser JN. Nod1
mediates cytoplasmic sensing of combinations of extracellular bacteria.
Cell Microbiol 2007;9:1343–51
23 Shuto T, Xu H, Wang B, Han J, Kai H, Gu X-X, Murphy TF, Lim DJ, Li
J-D. Activation of NF-kB by nontypeable Hemophilus influenzae is
mediated by toll-like receptor 2-TAK1-dependent NIK-IKKalpha/
b-IKBalpha and MKK3/6-p38 MAP kinase signaling pathways in
epithelial cells. Proc Natl Acad Sci USA 2001;98:8774–9
24 St. Geme JWS III, Falkow S, Barenkamp SJ. High-molecular-weight
proteins of nontypable Haemophilus influenzae mediate attachment to
human epithelial cells. Proc Natl Acad Sci USA 1993;90:2875–9
25 St Sauver J, Marrs CF, Foxman B, Somsel P, Madera R, Gilsdorf JR. Risk
factors for otitis media and carriage of multiple strains of Haemophilus
influenzae and Streptococcus pneumoniae. Emerg Infect Dis
2000;6:622–30
26 Swords WE, Buscher BA, Ver Steeg Ii K, Preston A, Nichols WA, Weiser
JN, Gibson BW, Apicella MA. Non-typeable Haemophilus influenzae
adhere to and invade human bronchial epithelial cells via an interaction
of lipooligosaccharide with the PAF receptor. Mol Microbiol
2000;37:13–27
27 McBroom AJ, Kuehn MJ. Release of outer membrane vesicles by
Gram-negative bacteria is a novel envelope stress response. Mol Microbiol
2007;63:545–58
28 Gu XX, Tsai CM, Apicella MA, Lim DJ. Quantitation and biological
properties of released and cell-bound lipooligosaccharides from
nontypeable Haemophilus influenzae. Infect Immun 1995;63:4115–20
29 Bjerre A, Brusletto B, Mollnes TE, Fritzsonn E, Rosenqvist E, Wedege E,
Namork E, Kierulf P, Brandtzaeg P. Complement activation induced by
purified Neisseria meningitidis lipopolysaccharide (LPS), outer membrane
vesicles, whole bacteria, and an LPS-free mutant. J Infect Dis
2002;185:220–8
30 Sharpe SW, Kuehn MJ, Mason KM. Elicitation of epithelial cell-derived
immune effectors by outer membrane vesicles of nontypeable
Haemophilus influenzae. Infect Immun 2011;79:4361–9
31 Tan TT, Morgelin M, Forsgren A, Riesbeck K. Haemophilus influenzae
survival during complement-mediated attacks is promoted by Moraxella
catarrhalis outer membrane vesicles. J Infect Dis 2007;195:1661–70
32 Kulp A, Kuehn MJ. Biological functions and biogenesis of secreted
bacterial outer membrane vesicles. Annu Rev Microbiol 2010;64:163–84
(Received November 3, 2011, Accepted January 7, 2012)
................................................................................................................................................
Ren et al. Long-term co-culture with NTHi
547