J Appl Physiol
91: 919–928, 2001.
Chronic airway infection leads to angiogenesis
in the pulmonary circulation
NATALIE HOPKINS, ELAINE CADOGAN, SHAY GILES, AND PAUL MCLOUGHLIN
Department of Human Anatomy and Physiology, Conway Institute of Biomolecular
and Biomedical Research, University College, Earlsfort Terrace, Dublin 2, Ireland
Received 31 August 2000; accepted in final form 10 April 2001
Pseudomonas aeruginosa; isolated-perfused lungs; stereology; pulmonary hypertension
and inflammation lead to remodeling of the pulmonary vasculature characterized
by an increased ratio of wall thickness to lumen diameter (2, 3, 7, 8, 26). This change in structure is similar
to that seen in pulmonary and systemic hypertension,
and it is commonly suggested that the thickened vessel
walls encroach on the vessel lumen, thus increasing
vascular resistance. However, we and others have reported that, in chronically infected lungs in rats, pulmonary hypertension did not occur and right ventricular hypertrophy was not observed, despite the
development of thickened vessel walls (2, 8). In a similar observation in human subjects with chronic obstructive pulmonary disease, Wright et al. (32) have
reported marked thickening of the walls of pulmonary
CHRONIC AIRWAY INFECTION
Address for reprint requests and other correspondence: P.
McLoughlin, Dept. of Human Anatomy and Physiology, Conway
Institute of Biomolecular and Biomedical Research, Univ. College,
Earlsfort Terrace, Dublin 2, Ireland (E-mail: paul.mcloughlin
@ucd.ie).
http://www.jap.org
arterial vessels in the absence of pulmonary hypertension.
One possible explanation of these findings is that
wall thickening occurred in an outward direction so
that the size of the vascular lumen was not compromised. This phenomenon, termed compensatory enlargement, has been reported previously in the systemic (7) but not in the pulmonary vasculature. A
second possible explanation is that angiogenesis occurs
in the pulmonary circulation in response to chronic
airway infection leading to an increase in the number
of parallel pathways through the lung, thus preventing
an increase in PVR. Angiogenesis in response to
chronic infection is well documented in the systemic
circulation, including the bronchial circulation (4, 5,
18). However, it is standard teaching that angiogenesis
does not occur in the adult pulmonary circulation (6,
10, 20).
The aim of the present study was to determine
whether chronic airway infection in rat lungs caused
angiogenesis or compensatory enlargement of the vessels in the pulmonary circulation. Such changes could
account for the absence of pulmonary hypertension in
these lungs, despite the development of vascular wall
thickening. We used a previously described model to
establish chronic airway infection with Pseudomonas
aeruginosa in rats (3, 8) and compared the structure of
lungs of chronically infected and control animals by
using quantitative stereological techniques. In addition, in isolated, ventilated, blood-perfused lungs, we
examined baseline pulmonary vascular resistance
(PVR) and changes in PVR produced by hypoxia.
METHODS
Infection of Animals
A mucoid Ps. aeruginosa strain, isolated from a patient
with cystic fibrosis, was used to prepare the inoculum for all
experiments. Chronic infection was produced by incorporating the organism into agarose beads as previously described
(2, 3, 8). In brief, a suspension of Ps. aeruginosa grown
overnight in peptone water resulted in a concentration of
⬃3 ⫻ 108 colony-forming units/ml. Nineteen milliliters of
agarose (2.1% wt/vol) were prepared and maintained at 45°C.
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
8750-7587/01 $5.00 Copyright © 2001 the American Physiological Society
919
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 16, 2017
Hopkins, Natalie, Elaine Cadogan, Shay Giles, and
Paul McLoughlin. Chronic airway infection leads to angiogenesis in the pulmonary circulation. J Appl Physiol 91:
919–928, 2001.—In both pulmonary and systemic hypertension, the walls of the arteriolar vessels are thickened and the
lumen size is reduced, leading to increased total vascular
resistance. It has been reported previously that chronic airway infection and inflammation lead to increased wall thickness in the pulmonary vasculature, without the development
of pulmonary hypertension. The aim of the present study was
to examine quantitatively the remodeling of intra-acinar
blood vessels in chronically infected rat lungs. Adult rats
were anesthetized and inoculated intratracheally with
Pseudomonas aeruginosa (n ⫽ 10) incorporated into agar
beads to induce chronic airway infection. Control groups
included rats inoculated with sterile agar beads (n ⫽ 8) and
rats that were not inoculated (n ⫽ 6). Chronic infection
caused vascular wall thickening without reduction in mean
lumen radius. Furthermore, chronic infection led to increased total length of intra-acinar vessels and increased
numbers of branch points, demonstrating that angiogenesis
had occurred. Preservation of lumen size and formation of
new parallel pathways in the vasculature of chronically infected lungs account for the maintenance of normal PVR
despite vessel wall remodeling.
920
INFLAMMATION AND PULMONARY ANGIOGENESIS
Lung Isolation for Hemodynamic Studies
Rats were anesthetized (60 mg/kg sodium pentobarbitone)
and mechanically ventilated (SAR-830P small animal ventilator, CWE, Ardmore, PA) at a tidal volume of 1.8 ml and a
frequency of 80 breaths/min. The animals were then anticoagulated (300 IU heparin intravenously) and killed by exsanguination. The thoracic contents were exposed through a
midline sternotomy, and cannulas were inserted into the
main pulmonary artery and left atrium and tied in place. The
thoracic contents were then removed en bloc and suspended
in a chamber maintained at 37°C while ventilation continued
with a warmed (37°C) and humidified mixture of 5% CO2 in
air. Airway pressure was continuously monitored, and a
positive end-expiratory pressure of 2 cmH2O was maintained. The lungs were briefly hyperinflated to an airway
pressure of 16 cmH2O every 5 min to prevent the development of progressive atelectasis.
The vascular perfusion circuit consisted of, in order, the
left atrial cannula, a venous outflow pressure transducer
(Sensor Nor 840, Horten, Norway), a warmed, thermostatically controlled, perfusate reservoir, a roller pump (Stockert
Instrumente, Munich, Germany), connecting tubing, bubble
trap, an arterial pressure transducer (Sensor Nor 840), and
the pulmonary artery cannula. The perfusion circuit was
primed with a mixture of 20 ml of rat blood, obtained by
exsanguination under general anesthesia (60 mg/kg sodium
pentobarbitone) of two anticoagulated (300 IU heparin) donor animals, and 10 ml physiological saline solution (in mM:
119 NaCl, 24 NaHCO3, 4.7 KCL, 0.9 MgSO4, 1.2 KH2PO4, 3.5
CaCl2, 5.5 glucose) with 4% (wt/vol) bovine serum albumin.
Perfusion was maintained at a constant flow (0.04
ml 䡠 mg⫺1 䡠 min⫺1) so that changes in arterial perfusion pressure reflected changes in total PVR. Venous outflow pressure
was maintained at 2 mmHg to ensure zone 3 conditions at
end expiration. All measurements of arterial perfusion presJ Appl Physiol • VOL
sure were made at end expiration. Arterial, venous, and
airway pressures were continuously recorded by using an
analog-to-digital system (Biopac MP100 WS, Linton Instrumentation, Norfolk, England) connected to a desktop computer (Power Macintosh 7100/80), and the data were stored
on hard disc for later analysis.
Hemodynamic Studies
After isolation, a period of equilibration was allowed until
airway and vascular pressures were stable while the lungs
were ventilated with normoxic gas (a mixture of 5% CO2 and
95% air). A hypoxic challenge was presented by switching the
ventilating gas to a mixture of 3% O2, 5% CO2, and 92% N2
for 10 min. Hemodynamic measurements were made at the
end of this period. Ventilation with normoxic gas was then
resumed, and perfusion pressure was allowed to return to
baseline values. Perfusate was sampled regularly, and its
pH, PCO2, and PO2 were measured by using an automatically
calibrating blood gas analyzer (Ciba Corning Model 278,
Medfield, MA). pH was maintained in the range 7.38–7.44
under all conditions by addition of 0.1 M NaHCO3 as required. The perfusion protocol lasted ⬃90 min in total.
Bronchoalveolar Lavage
Immediately after the perfusion protocol, bronchoalveolar
lavage (BAL) was carried out by intratracheal instillation of
four separate aliquots of normal saline (5 ml) and collection
of the returned fluid by free drainage. Total cell numbers per
milliliter in the BAL fluid were counted, and differential cell
counts were performed after staining with Diff-Quick (Dade).
The pulmonary circulation was then perfused with calciumfree normal saline at 37°C until the effluent was clear of
blood. Lungs were fixed for morphometric examination by the
technique of Meyrick and Reid (24). Paraformaldehyde (4%
wt/vol) in PBS (300 mosM) at a pressure of 100 cmH2O and a
temperature of 37°C was instilled through the pulmonary
artery catheter to maximally distend the pulmonary vessels.
During fixation, the lungs were simultaneously inflated
through the tracheal catheter using the same fixative at a
pressure of 25 cmH2O (24). After this, the pulmonary artery
and trachea were ligated and the lungs were then stored in
fixative until they were embedded in paraffin wax.
Measurement of Right Ventricular Weights
The rat hearts were isolated along with the lungs. They
were then excised and placed in fixative (4% wt/vol paraformaldehyde in PBS). The atria were removed, the right ventricular free wall (RV) was dissected free of the left ventricle
and septum (LV ⫹ S), and each ventricle was weighed separately. The results were then expressed as milligrams per
100 g body wt. The ratio of RV to LV ⫹ S weight was
calculated as an index of RV hypertrophy induced by increases in PVR relative to systemic vascular resistance.
Morphometric Measurements
Estimation of lung volume and volumes of pulmonary
tissue compartments. The vertical axis of each left lung was
identified, and the lung was cut perpendicular to this axis
into slices (4 mm thick) with a sharp blade beginning at a
position chosen by random number in the first slice. The
modal number of slices obtained was 7 (range 6–8). To
determine the volume of the lung, an image of the surface
area of each slice was obtained by using a JVC KY-F55B color
video camera (Eurotek, Dublin, Ireland). The images were
digitized and displayed on screen using Adobe Photoshop,
91 • AUGUST 2001 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 16, 2017
To this, 1 ml of the peptone broth was added, and the
resultant agarose solution was injected into 20 ml of heated
(50°C) mineral oil (Sigma Chemical) and mixed vigorously for
a further 10 min. The mixture was then cooled rapidly by
immersing the beaker in crushed ice while stirring continued, leading to the formation of agar beads. The beads were
separated by centrifugation, and residual mineral oil was
removed by washing in 5% (wt/vol) sodium deoxycholate
solution in PBS (0.1 M) followed by a second wash in a 0.25%
(wt/vol) solution and subsequently washed four times in PBS.
Finally, the beads were resuspended in an equal volume of
PBS for inoculation.
Adult male (300–400 g) specific pathogen-free SpragueDawley rats (Harlan, Bicester, UK) were anesthetized (Hypnorm: fentanyl citrate 0.25 mg/kg and fluanisone 0.08 mg/kg,
midazolam 2.5 mg/kg sc), and a modified pediatric laryngoscope was used to introduce a polyethylene cannula (1.2-mm
outside diameter) into the trachea via the larynx. In the
group to be chronically infected, 104 colony-forming units of
Ps. aeruginosa in agar beads suspended in PBS (total volume
200 ␮l) were inoculated intratracheally through this cannula
in each rat, and the animals were then allowed to recover
from anesthesia. A second group of animals (placebo-inoculated group) was anesthetized and inoculated with sterile
agarose beads, that is, agar beads prepared in a similar
manner except that Pseudomonas organisms were omitted. A
third group consisted of animals that were not inoculated
(noninoculated group). Isolation of lungs for hemodynamic,
histological, and immunohistochemical analyses was carried
out 10–15 days postinoculation.
INFLAMMATION AND PULMONARY ANGIOGENESIS
J Appl Physiol • VOL
tunica adventitia was defined as the loose connective tissue
investing the vessel external to the external elastic lamina.
Estimation of the length of intra-acinar blood vessels. To
estimate the length density of the intra-acinar pulmonary
blood vessels of each lung, the number of intra-acinar blood
vessels, which transected counting frames randomly superimposed on isotropic uniform random sections, was counted
(1, 11, 12). The modal number of fields of view examined in a
lung was 24 (range 22–34).
Estimation of number of branch points of intra-acinar
pulmonary blood vessels. The number of branch points of
intra-acinar pulmonary blood vessels was determined using
the physical double dissector (1, 11, 12). A branch point was
defined as where two immediately adjacent vessels shared a
common wall (Fig. 1). Vertically aligned images separated by
14 ␮m were examined by using a systematic random sampling strategy. The modal number of dissector pairs examined in a lung was 72 (range 66–102).
The number of fields of view from each lung that were
examined differed for estimating volume, length, and numerical densities of the specific compartments of interest, but for
all parameters equal numbers of fields of view from each
block were examined. The counting strategies (number of
tissue blocks and fields of view) were devised in preliminary
studies so that the number of counts of the item of interest
(point hits on a specific compartment, vessel intersections or
counts on dissector pairs) was never ⬍100–200 in any single
lung. Increasing the number of counts above this range does
not significantly improve the precision of the estimate obtained (11–13, 31). This aim was successfully achieved in all
estimates of volume, numerical, and length densities except
that of the media of intra-acinar vessels in control lungs. In
the case of the volume density of the media of intra-acinar
vessels, point counts in the chronically infected lungs exceeded 600 in all lungs (mean number 983 ⫾ 67) but were
⬍100 in all lungs of both control groups (mean number 24 ⫾
7). However, because the differences between the groups
Fig. 1. Photomicrograph of noninoculated lung tissue showing a
typical branch point of an intra-acinar pulmonary blood vessel (v).
Scale bars ⫽ 20 ␮m.
91 • AUGUST 2001 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 16, 2017
stored in eight-bit (256 level) format, and then imported into
Stereology Toolbox (Morphometrix) for determination of surface area by use of a point-counting grid. Lung volume was
than calculated by the Cavalieri method (1, 11, 12). The lung
slices were cut into bars of tissue and every third bar was
selected, beginning at a point identified by randomly choosing a number between one and three. The selected bars were
cut into blocks using a vertical uniform random strategy,
with cuts made at 2-mm intervals (1). Every fourth block was
selected for embedding beginning at a start point identified
by randomly choosing a number between one and four. The
modal number of blocks selected from a single lung was 12
(range 11–17). The blocks were embedded in wax, and random sections (7 ␮m thick) were cut. To obtain isotropic
uniform random sections, random blocks (one in four, as
above) were initially cut in a randomly chosen direction
about the vertical axis and then cut in a random direction
chosen using a cosine-weighted orientator clock that was
projected onto the block face exposed by the initial vertical
uniform random cut (22). Sections (7 ␮m thick) were cut
parallel to this direction and randomly selected sections were
stained with an elastin stain (Miller’s) and neutral red.
Fields of view from each section were examined by light
microscopy (Leica, Laboratory Instruments), captured using
the JVC KY-F55B color video camera mounted on the camera
port of the microscope. The images were digitized and displayed on screen and imported into Stereology Toolbox for
analysis (see above).
Volume densities of specific tissues in the lung were estimated by point counting, and the absolute volume was calculated using the lung volumes measured by the Cavalieri
method (1, 11, 12).
To determine the volumes of the vascular compartments of
interest, a cascade approach was used (1, 11, 12). The volume
densities of intra- and extra-acinar tissues (including all
air-spaces) were initially estimated by point counting. Intraacinar blood vessels were defined as those associated with
respiratory bronchioles, alveolar ducts, alveolar sacs, and
alveoli, i.e., vessels within the gas exchange region of the
lung. The modal number of randomly chosen fields of view
examined was 12 (range 11–17), one field from a randomly
chosen section from each block. The volume densities of the
intra-acinar pulmonary blood vessels excluding capillaries,
alveolar walls, and intra-acinar air-spaces per unit volume of
intra-acinar tissue were then measured by point counting on
random fields of view obtained at higher magnification. The
modal number of randomly chosen fields of view examined
was 24 per lung (range 22–34), two from each of 12 randomly
chosen sections from each block. To estimate the volumes of
the vessel lumen, the tunica intima, tunica media, and tunica
adventitia within each left lung, images of randomly selected
intra-acinar vessels were acquired and placed randomly
within point-counting grids. To randomly sample intra-acinar vessels, the following strategy was adopted: a search for
vessels was begun at one of the four corners of the section
that had been chosen randomly. Contiguous square fields of
view were sequentially examined by sweeping across the
section horizontally until the opposite edge was reached,
then moving the section vertically by a distance equal to that
of one field of view, and then returning across the section.
This pattern was continued until the five vessels first encountered in each block were examined. The modal number of
vessels examined in a lung was 60 (range 55–85). The tunica
intima was defined as the internal elastic lamina and the
cells of the endothelium internal to it. The tunica media was
defined as the external elastic lamina and everything internal to it but external to the internal elastic lamina. The
921
922
INFLAMMATION AND PULMONARY ANGIOGENESIS
were large in this particular case, adequate precision of the
estimates was obtained.
Calculation of vessel radius and the ratio of blood vessel
wall thickness to vessel radius. The vascular smooth muscle
was fully relaxed by perfusion with calcium-free solution and
then fixed in a fully distended state by infusion of fixative at
high pressure, as already described. Thus the vessels were
assumed to be cylindrical in shape, and the radius of the
vessel was calculated from the standard formula for the
volume of a cylinder and the known volume of the intraacinar vessels (volume-derived radius). The ratio of the intraacinar blood vessel wall thickness to the radius of the lumen
was calculated as
WT/vessel radius ⫽ 兵关V共media兲 ⫹ V共intima兲 ⫹ V共lumen兲兴0.5
Table 2. Total and differential cell counts in BAL
fluid from noninoculated, Pseudomonas-inoculated,
and placebo-inoculated lungs
Total cell count, ⫻107
cells/ml
Neutrophils, %
Macrophages, %
Lymphocytes, %
Noninoculated
(n ⫽ 8)
PseudomonasInoculated
(n ⫽ 10)
PlaceboInoculated
(n ⫽ 6)
4.2 ⫾ 2.2
2.7 ⫾ 0.6
97.0 ⫾ 0.8
0.52 ⫾ 0.8
17.4 ⫾ 12.0*
20.0 ⫾ 8.0*
72.0 ⫾ 13.0*
9.0 ⫾ 4.0*
4.1 ⫾ 1.1
2.8 ⫾ 0.7
96.0 ⫾ 1.4
1.2 ⫾ 0.8
Values are means ⫾ SE. * Significant difference from noninoculated and placebo-inoculated groups (P ⬍ 0.01, ANOVA).
⫺ 关V共lumen兲兴0.5其/关V共media兲 ⫹ V共intima兲 ⫹ V共lumen兲兴0.5
where WT is the blood vessel wall thickness measured from
the innermost aspect of the tunica intima to the outside of the
external elastic lamina.
Values are expressed as means ⫾ SE. Volumes of specific
tissue compartments, vessel lengths, and numbers of branch
points in the left lung are reported per 100 g body wt.
Statistical comparisons of means were made using ANOVA,
and, when this indicated significant differences between
groups, the Student-Newman-Keuls post hoc test was used to
assess the significance of the differences between specific
means. Where appropriate (total and differential cell counts),
data were log transformed before ANOVA. For clarity, the
untransformed values are presented in the text. A value of
P ⬍ 0.05 was accepted as statistically significant.
RESULTS
There were no significant differences in mean body
weight and mean right ventricular weight between the
three groups of animals under study (Table 1). Mucoid
colonies of Ps. aeruginosa were grown on blood agar
plates from the BAL fluid obtained from each chronically infected lung. Organisms were not isolated from
either the noninoculated or the placebo-inoculated
lungs. Differential cell counts showed that the mean
percentage of neutrophils and lymphocytes in the
Pseudomonas-infected lungs was significantly elevated
above that in the other two groups (Table 2). The mean
total cell count in the BAL fluid from PseudomonasTable 1. Body weight, RV weight, RV/LV ratio, and
hematocrit in noninoculated, Pseudomonasinoculated, and placebo-inoculated animals
Body weight, g
RV weight, mg/100 g
body wt
RV/LV ratio
Hematocrit, %
Noninoculated
(n ⫽ 8)
PseudomonasInoculated
(n ⫽ 10)
PlaceboInoculated
(n ⫽ 6)
362 ⫾ 15.0
348 ⫾ 8.0
369 ⫾ 15.0
48.0 ⫾ 4.0
0.16 ⫾ 0.01
44.0 ⫾ 0.7
51.0 ⫾ 7.0
0.18 ⫾ 0.03
45.0 ⫾ 0.6
51.0 ⫾ 6.9
0.18 ⫾ 0.03
44.0 ⫾ 0.5
Values are means ⫾ SE. RV, right ventricular free wall RV/LV
ratio, ratio of weight of the RV to left ventricular and septal weight.
No significant differences between group mean values were observed.
J Appl Physiol • VOL
Lung Histology
Lungs from noninoculated animals showed normal
alveolar structure with no evidence of inflammatory
cell infiltration (Fig. 2A). In contrast, the Pseudomonas-inoculated group showed evidence of extensive regions of chronic inflammation with extensive thickening of the alveolar walls due to the infiltration of
inflammatory cells (Fig. 2B). Markedly increased numbers of inflammatory cells were observed in the alveolar walls of the Pseudomonas-inoculated group (Fig. 2,
B and E). However, these inflammatory changes
were not homogeneously distributed throughout the
lung, so that some areas within an infected lung
were relatively normal in structure (Fig. 2C),
whereas others showed extensive inflammatory
changes (Fig. 2, B and E). The placebo-inoculated
group had normal alveolar structure without evidence of inflammation (Fig. 2D).
In the noninoculated group, the vessel wall of intraacinar pulmonary blood vessels generally contained a
single elastic lamina (Fig. 3A), although occasional
vessels were seen with a double elastic lamina. In
contrast, the Pseudomonas-inoculated group showed
frequent thickening of the tunica media, with internal
and external elastic laminas (Fig. 3B). The extent of
the remodeling varied throughout the lungs with some
vessels showing lesser medial thickening and still
Table 3. Baseline and change in Ppa
in response to hypoxia
Baseline Ppa, mmHg
Hypoxia, ⌬Ppa,
mmHg
Noninoculated
(n ⫽ 8)
PseudomonasInoculated
(n ⫽ 10)
PlaceboInoculated
(n ⫽ 6)
16.0 ⫾ 1.2
15.3 ⫾ 1.1
17.3 ⫾ 1.5
23.4 ⫾ 1.2
27.7 ⫾ 4.2
27.5 ⫾ 3.9
Values are means ⫾ SE. Ppa, pulmonary arterial pressure;
⌬Ppa ⫽ change in Ppa. No significant differences between group
mean values were observed.
91 • AUGUST 2001 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 16, 2017
Data Analysis
infected lungs was significantly greater than that in
the noninoculated and placebo-inoculated groups (Table 2). Mean baseline perfusion pressure and the mean
hypoxic vasoconstrictor response were similar in the
three groups of isolated lungs (Table 3).
INFLAMMATION AND PULMONARY ANGIOGENESIS
923
other vessels having a normal structure. In general,
vessel wall remodeling was most extensive in the most
inflamed regions of lung, whereas in regions of lung
that were not inflamed vessel structure was normal.
The placebo-inoculated group showed vessels that predominately contained a single elastic lamina (Fig. 3C),
similar to those seen in control lungs.
J Appl Physiol • VOL
Morphometric Analysis
Volumes of pulmonary tissue compartments. Mean
total lung volume of the Pseudomonas-inoculated
group was significantly (P ⬍ 0.05) greater than those
from both the noninoculated and placebo-inoculated
groups (Fig. 4), whereas both control groups, the non-
91 • AUGUST 2001 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 16, 2017
Fig. 2. A: photomicrograph of noninoculated lung tissue demonstrating
normal alveolar structure (Miller’s
stain). B: photomicrograph of Pseudomonas-inoculated lung tissue demonstrating thickened and distorted alveolar walls due to inflammatory cell
infiltration (Miller’s stain). C: photomicrograph of a region from a
chronically infected lung (same lung
shown in B) demonstrating relatively
normal alveolar structure and absence of inflammatory cell infiltrate
(Miller’s stain). D: photomicrograph of
placebo-inoculated lung showing normal alveolar structure (Miller’s stain).
E: photomicrograph of Pseudomonasinoculated lung tissue stained with
hematoxylin and eosin demonstrating
thickened alveolar walls and inflammatory cell infiltrate. Scale bars ⫽ 20
␮m in all panels.
924
INFLAMMATION AND PULMONARY ANGIOGENESIS
inoculated and placebo-inoculated, did not differ significantly from one another. Extra- and intra-acinar tissue volumes were significantly greater in the
Pseudomonas-inoculated group than in either of the
two control groups (Fig. 4).
Mean total alveolar wall volume was significantly
greater in the chronically infected group than in both
control groups. There was also a significant difference
between the mean volume of intra-acinar pulmonary
blood vessels in the infected and both control groups,
whereas no significant differences were observed in
air-space volumes between the three groups. These
results indicate that the increase in intra-acinar tissue
volume was due to the increase in the volume of alveolar walls and intra-acinar blood vessels.
Volumes of the tissue compartments of intra-acinar
blood vessels. The mean total volume of the tunica intima, tunica media, and tunica adventitia of the intraacinar pulmonary vessels showed significant increases in
the Pseudomonas-inoculated group when compared with
noninoculated and placebo-inoculated groups (Fig. 5).
Length of intra-acinar pulmonary vessels. The mean
total length of the intra-acinar pulmonary blood vessels per left lung was significantly greater in the
Pseudomonas-inoculated lungs than in the noninoculated and placebo-inoculated groups (Fig. 6).
Number of branch points in intra-acinar pulmonary
vessels. The mean total number of branch points in the
intra-acinar pulmonary blood vessels per left lung was
Fig. 3. A: photomicrograph showing normal intra-acinar pulmonary
blood vessel (v) from noninoculated lung. B: photomicrograph of
Pseudomonas-inoculated lung tissue showing remodeled intra-acinar pulmonary blood vessel (v). C: photomicrograph of an intraacinar pulmonary blood vessel (v) from a placebo-inoculated lung.
Scale bar ⫽ 40 ␮m in all panels.
J Appl Physiol • VOL
91 • AUGUST 2001 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 16, 2017
Fig. 4. Mean ⫾ SE volumes of left lungs and subcompartments in
noninoculated (n ⫽ 8), Pseudomonas-inoculated (n ⫽ 10), and placebo-inoculated (n ⫽ 6) animals. *Significant difference from noninoculated and placebo-inoculated groups (P ⬍ 0.05, ANOVA). V(lung),
volume of left lung; V(ac), volume of intra-acinar tissue; V(ex),
volume of extra-acinar tissue; V(a.sp.), volume of intra-acinar air
space; V(vsl), volume of intra-acinar pulmonary blood vessels excluding capillaries; V(al.wall), volume of alveolar wall.
925
INFLAMMATION AND PULMONARY ANGIOGENESIS
significantly greater in the Pseudomonas-inoculated
lungs than in the noninoculated and placebo-inoculated groups (Fig. 7).
Blood vessel wall thickness and lumen radius. The
mean ratio of blood vessel wall thickness to vessel
radius was significantly greater (P ⬍ 0.05) in the infected (0.31 ⫾ 0.01) than in the noninoculated (0.12 ⫾
0.02) and placebo-inoculated (0.14 ⫾ 0.02) groups.
However, the mean diameter of the vascular lumen in
the infected lungs (36 ⫾ 1.3 ␮m) was not significantly
different from that in the placebo (37 ⫾ 1.4 ␮m) and
noninoculated (42 ⫾ 2.8 ␮m) groups. The mean thick-
Fig. 7. Mean ⫾ SE number of branch points of intra-acinar pulmonary blood vessels in the left lungs of the noninoculated (n ⫽ 8),
Pseudomonas-inoculated (n ⫽ 10), and placebo-inoculated (n ⫽ 6)
animals. *Significant difference from noninoculated and placeboinoculated groups (P ⬍ 0.05, ANOVA).
ness of the internal elastic lamina and the intima
internal to it in the chronically infected lungs was
3.4 ⫾ 0.2 ␮m, a value not significantly different from
that of the placebo-inoculated (3.0 ⫾ 0.5 ␮m) and of the
noninoculated (2.6 ⫾ 0.5 ␮m) groups.
The increased volumes of each of the three layers of
the vessel wall that were observed may have arisen
because of the increased total length alone. Thus we
computed the volume of each compartment per unit
vessel length; results are shown in Table 4. Only in the
case of the tunica media was a significant increase in
volume per unit length observed, demonstrating that
this is the layer that became thickened as a result of
chronic infection.
DISCUSSION
We demonstrated that chronic airway infection leads
to vascular remodeling in the pulmonary circulation,
i.e., a thickening of the vessel walls compared with the
radius of the vessel. Despite these changes in vessel
Table 4. Volumes of tunica intima, tunica media, and
tunica adventitia in intra-acinar pulmonary blood
vessels per unit length of blood vessel
PseudomonasInoculated
(n ⫽ 10)
PlaceboInoculated
(n ⫽ 6)
3.5 ⫾ 0.8
4.3 ⫾ 0.4
3.8 ⫾ 0.8
0.2 ⫾ 0.08
6.9 ⫾ 0.8*
0.2 ⫾ 0.9
8.6 ⫾ 0.8
8.4 ⫾ 0.6
7.3 ⫾ 0.8
Noninoculated
(n ⫽ 8)
Fig. 6. Mean ⫾ SE total length of intra-acinar pulmonary blood
vessels in the left lungs of the noninoculated (n ⫽ 8), Pseudomonasinoculated (n ⫽ 10), and placebo-inoculated (n ⫽ 6) animals. *Significant difference from noninoculated and placebo inoculated groups
(P ⬍ 0.05, ANOVA).
J Appl Physiol • VOL
Volume of intima per
unit length, ⫻10⫺7
cm3/cm
Volume of media per
unit length, ⫻10⫺7
cm3/cm
Volume of adventitia
per unit length,
⫻10⫺7 cm3/cm
Values are means ⫾ SE. * Significant difference from noninoculated and placebo-inoculated groups (P ⬍ 0.01, ANOVA).
91 • AUGUST 2001 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 16, 2017
Fig. 5. Mean ⫾ SE volumes of lumen and tunics of the intra-acinar
pulmonary blood vessels in the left lungs of the noninoculated (n ⫽
8), Pseudomonas-inoculated (n ⫽ 10), and placebo-inoculated (n ⫽ 6)
animals. V(lumen), volume of intra-acinar blood vessel lumen; V(intima), volume of tunica intima; V(media), volume of tunica media;
V(adventitia), volume of tunica adventitia. *Significant difference
from noninoculated and placebo-inoculated groups (P ⬍ 0.01,
ANOVA).
926
INFLAMMATION AND PULMONARY ANGIOGENESIS
J Appl Physiol • VOL
tional to the increase in vessel length, implying that
neither of these two tunics was thickened relative to
the size of the blood vessels. However, the calculated
mean lumen radius in the chronically infected group
was not significantly reduced compared with the two
control groups, suggesting that thickening of the vessel
wall does not inevitably lead to encroachment on, and
reduction of, the lumen. Rather, the enlargement of the
wall occurred predominantly in an outward direction,
causing an increase in the external diameter of the
vessel. A similar behavior has been reported previously
in atherosclerotic disease of systemic arteries, in which
the vessel enlarged in an outward direction at the site
of an atherosclerotic plaque so that the lumen size
remained normal (7). This phenomenon has not previously been reported in the pulmonary circulation. Our
findings also demonstrate that medial hypertrophy
does not necessarily imply increased vascular reactivity, because hypoxic vasoconstriction was not augmented in the chronically infected lungs (Table 3).
There are a number of possible explanations for this
observation. Vascular remodeling was not uniform
throughout all vessels in the chronically infected lungs,
and a large number of vessels were observed in which
wall structure was relatively normal (see Lung Histology and Fig. 2). These normal vessels may have provided a low-resistance pathway through the lung that
allowed normal pulmonary artery pressure to be preserved. Such preservation of normal pulmonary artery
pressure would be in keeping with previous reports
that pulmonary artery pressure remains normal after
pneumonectomy, i.e., after loss of half the pulmonary
vascular bed (21). A further potential explanation is
that the newly developed medial smooth muscle was
predominantly proliferative in phenotype and therefore had poor contractile function (29). We did not
examine these possibilities in the present study, and
further work is needed to elucidate this issue.
The present results also show a substantial increase
in total intra-acinar vessel length (Fig. 6). If this occurred because of increased length and tortuosity of
existing vessels, then PVR would have risen, because
mean lumen radius remained unchanged. However, if
the increase in vessel length resulted, at least in part,
from the formation of new vessels that ran in parallel
to the preexisting pathways, then PVR could be maintained at normal values. Our finding that there were a
significantly increased number of branch points in the
pulmonary circulation supports the hypothesis that
angiogenesis occurred as a result of chronic airway
infection (Fig. 7). The relative increase in the number
of branch points in infected lungs appears somewhat
greater than that of vessel length (Figs. 6 and 7). This
finding suggests that the mean length of new vessels
may be shorter than that of the preexisting vessels.
This evidence of angiogenesis in the pulmonary circulation is in contrast to previous reports that neither
chronic hypoxia (10, 19, 30) nor chronic inflammation
(27, 28) causes new vessel formation in this circulation.
Both of these stimuli are powerful initiators of angiogenesis in the systemic circulation (9, 20). It is inter-
91 • AUGUST 2001 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 16, 2017
structure, pulmonary hypertension did not develop, as
demonstrated by the absence of right ventricular hypertrophy (Table 1) and normal PVR in isolated lungs
from chronically infected animals (Table 3). This absence of pulmonary hypertension is in agreement with
previous reports of chronic airway infection in rats,
which demonstrate, after a similar time interval, structural changes in the vascular wall, including medial
thickening, without increases in PVR (2, 8).
The structure of the pulmonary vasculature of the
rat has been extensively studied previously, most notably by Reid and colleagues (15, 19, 24). They report
that intra-acinar arterial vessels vary in external diameter from 15 to 150 ␮m in the rat (15). Given that
the smaller vessels are more numerous than larger
ones, our volume-averaged lumen diameter of intraacinar vessels in noninoculated and placebo-inoculated
control lungs is in good agreement with the range of
vessel diameters reported previously (15). Our estimate of the mean luminal diameter of intra-acinar
vessels also agrees well with that obtained from previously reported models of the pulmonary arterial tree
(14, 16, 17, 33). The mean luminal diameter depends
on the relative numbers of vessels of different diameters and their lengths. Haworth et al. (14) showed that
the slope of the power law relationship between the
numbers of intrapulmonary vessels and their diameters and between the segment lengths and their diameters is relatively constant across species and can be
used to determine the numbers and dimensions of the
pulmonary vessels. We used the model of Haworth et
al. to determine the mean volume-derived diameter of
the intra-acinar vessels. Assuming that the mean diameter of the immediate precapillary arterioles is 15
␮m and that the intra-acinar vessels are not more than
150 ␮m in diameter in the rat (24), then the volumederived mean diameter of the intra-acinar vessels predicted by the model is 35 ␮m; this value is in good
agreement with that which we observed in control
tissues. Using electron microscopy, Meyrick and Reid
(24) found that the internal elastic lamina is up to 1.4
␮m thick and that the mean thickness of intima internal to this ranges between 0.2 and 4.0 ␮m. This suggests that the thickness of the two compartments combined ranged from slightly greater than 1 ␮m to 5.4 ␮m
along the intra-acinar vessels. On the basis of stereological analysis, we estimated that the thickness of
internal elastic lamina, the endothelium, and other
cells internal to the internal elastic lamina in the
control noninoculated lungs, was 2.5 ⫾ 0.5 ␮m, a value
in good agreement with the data of Meyrick and Reid.
Taken together, these observations indicate that the
stereological approach that we used provided information about intra-acinar vessel structure compatible
with previous studies using other techniques.
In chronically infected lungs, we found increased
medial volume per unit length of intra-acinar vessel
wall (Table 4), a finding that agrees with those based
on measurements of wall thickness and lumen diameter in histological sections (2, 3, 8). The increase in
intimal and adventitial volumes observed was propor-
INFLAMMATION AND PULMONARY ANGIOGENESIS
J Appl Physiol • VOL
maintain lumen cross-sectional area and provide new
vascular pathways through the lung, which prevent
the development of pulmonary hypertension despite
the development of significant thickening of the vessel
wall. Our findings also suggest that pulmonary angiogenesis may play an important role in lung disease,
including neoplastic and chronic inflammatory disorders.
We thank D. Briton, St. Vincents Hospital, for help and advice in
preparing the Pseudomonas inoculum.
This work was supported by the Health Research Board of Ireland.
REFERENCES
1. Bolender RP, Hyde DM, and Dehoff RT. Lung morphometry:
a new generation of tools and experiments for organ, tissue, cell,
and molecular biology. Am J Physiol Lung Cell Mol Physiol 265:
L521–L548, 1993.
2. Cadogan E, Hopkins N, Giles S, Bannigan J, Moynihan J,
and McLoughlin P. Enhanced expression of inducible nitric
oxide synthase without vasodilator effect in chronically infected
lung. Am J Physiol Lung Cell Mol Physiol 277: L616–L627, 1999.
3. Cash HA, Woods DE, McCullough B, Johanson WG, and
Bass JA. A rat model of chronic respiratory infection with
Pseudomonas aeruginosa. Am Rev Respir Dis 119: 453–459,
1978.
4. Charan NB and Carvalho PG. The bronchial circulation in
chronic lung infections. In: The Bronchial Circulation, edited by
Butler J. New York: Marcel Dekker, 1992, p. 535–549.
5. Cudkowicz L. Bronchial arterial circulation in man. In: Pulmonary Vascular Disease, edited by Moser KM. New York: Marcel
Dekker, 1979, p. 111–232.
6. DeGraff AC Jr, Grover RF, Johnson RL Jr, Hammond JW
Jr, and JM Miller. Diffusing capacity of the lung in Caucasians
native to 3,100 m. J Appl Physiol 29: 71–76, 1970.
7. Glagov S, Weisenberg E, Zarins CK, Stankunavicius R,
and Kolettis GJ. Compensatory enlargement of human atherosclerotic coronary arteries. N Engl J Med 22: 1371–1375, 1993.
8. Graham LM, Vasil A, Vasil ML, Voelkel NF, and Stenmark
KR. Decreased pulmonary vasoreactivity in an animal model of
chronic Pseudomonas pneumonia. Am Rev Respir Dis 142: 221–
229, 1990.
9. Griffioen AW and Molema G. Angiogenesis: potentials for
pharmacologic intervention in the treatment of cancer, cardiovascular diseases, and chronic inflammation. Pharmacol Rev 52:
237–268, 2000.
10. Grover RF, Wagner WW, McMurtry IF, and Reeves JT.
Pulmonary circulation. In: Handbook of Physiology. The Cardiovascular System. Peripheral Circulation and Organ Blood Flow.
Bethesda, MD: Am. Physiol. Soc., 1983, sect. 2, vol. III, pt. 1,
chapt. 4, p. 103–136.
11. Gundersen HJ. Stereology of arbitrary particles. A review of
unbiased number and size estimators and the presentation of
some new ones, in memory of William R. Thompson. J Microsc
143: 3–45, 1986.
12. Gundersen HJG, Bagger P, Bendtsen TF, Evans TF, Korbo
L, Marcussen N, Moller A, Nielsen K, Nyengaard JR, Pakkenberg B, Sorensen FB, Vesterby A, and West MJ. The
new stereological tools: disector, fractionator, nucleator and
point sampled intercepts and their use in pathological research
and diagnosis. APMIS 96: 857–881, 1988.
13. Gundersen HJG, Bendtsen TF, Korbo L, Marcussen N,
Moller A, Nielsen K, Nyengaard JR, Pakkenberg B, Sorensen FB, Vesterby A, and West MJ. Some new simple and
efficient stereological methods and their use in pathological
research and diagnosis. APMIS 96: 379–394, 1988.
14. Haworth ST, Linehan JH, Bronikowski TA, and Dawson
CA. A hemodynamic model representation of the dog lung.
J Appl Physiol 70: 15–26, 1991.
15. Hislop A and Reid L. Normal structure and dimensions of the
pulmonary arteries in the rat. J Anat 125: 71–83, 1978.
91 • AUGUST 2001 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 16, 2017
esting to note that, in the setting of metastatic lung
disease, Milne and Zerhouni (25) have reported that
most such tumors in the lung receive all, or a major
part, of their blood supply from the pulmonary circulation.
It is likely that the explanation for these differing
reports lies in the difficulty of identifying new vessels
in an organ that is as vascular as the lung. When many
blood vessels are already present in a highly vascular
tissue, the addition of a small number of new vessels is
very difficult to detect by direct inspection of twodimensional sections, the method used in previous
studies (27, 28, 30). Furthermore, counting the number
of vessels viewed per high-power field will also fail to
detect change in total vessel length, because such an
approach does not allow for the effects of changes in the
volume of the reference space, e.g., the size of the lung
or the size of the alveolus. However, the techniques of
stereological morphometry circumvent these difficulties and allowed us to obtain quantitative information
on the three-dimensional structure of the lung vasculature demonstrating increased total length and new
branch-point formation. Neither of these changes could
have been detected by simple inspection of random
histological sections.
The role played by vascular remodeling in the development of pulmonary hypertension in chronic lung
infection is complex. Our data suggest that, under
certain conditions, wall remodeling may be not result
in reduction of the mean lumen diameter. Furthermore, the addition of parallel vascular pathways
through the lungs may serve to oppose any increase in
total PVR that might result from reductions in vascular lumen diameter in some vessels. When these two
processes occur together, vascular resistance would not
increase despite thickening of the vessel wall. However, in other circumstances, such as more persistent
or more severe infections, remodeling of the vascular
wall, including the wall of newly formed vessels, might
lead to significant reductions in mean lumen diameter.
Under such conditions, the increase in resistance
caused by narrowed vessels might outweigh the extent
to which new vessel formation could reduce resistance
and the net effect might then become an increase in
total PVR. Graham et al. (8) found that, in chronic
Pseudomonas infection of rat lungs, pulmonary hypertension was not observed 2 wk after infection but was
present after 6–9 wk, suggesting that more prolonged
inflammation caused increased vascular resistance.
However, the duration of infection is clearly not the
only factor that influences the development of hypertension. McCormack and Paterson (23) found significantly increased PVR 7–10 days after intratracheal
inoculation of Pseudomonas incorporated into agar
beads. This suggests that other influences, such as
severity of infection, virulence of the infecting organism, and host factors, also play important roles.
In conclusion, we have demonstrated, for the first
time, that chronic airway infection may cause pulmonary vascular remodeling together with angiogenesis
in the intra-acinar vasculature. These changes can
927
928
INFLAMMATION AND PULMONARY ANGIOGENESIS
J Appl Physiol • VOL
26. Peinado VI, Barbera JA, Abate P, Ramirez J, Roca J,
Santos S, and Rodriguez-Roisin R. Inflammatory reaction in
pulmonary muscular arteries of patients with mild chronic obstructive pulmonary disease. Am J Respir Crit Care Med 159:
1605–1611, 1999.
27. Schraufnagel DE. Monocrotaline-induced angiogenesis. Differences in the bronchial and pulmonary vasculature. Am J Pathol
137: 1083–1090, 1990.
28. Schraufnagel DE, Mehta D, Harshbarger R, Treviranus K,
and Wang NS. Capillary remodeling in bleomycin-induced pulmonary fibrosis. Am J Pathol 125: 97–106, 1986.
29. Schwartz SM, Campbell GR, and Campbell JH. Replication
of smooth muscle cells in vascular disease. Circ Res 58: 427–444,
1986.
30. Sobin SS, Tremer HM, Hardy JD, and Chiodi HP. Changes
in arteriole in acute and chronic hypoxic pulmonary hypertension and recovery in rat. J Appl Physiol 55: 1445–1455, 1983.
31. Weibel ER and Cruz-Orive LM. Morphometric methods. In:
The Lung: Scientific Foundations (2nd ed.), edited by Crystal
RG, West JB, Weibel ER, and Barnes PJ. Philadelphia, PA:
Lippincott-Raven, 1997, p. 333–344.
32. Wright J, Lawson L, Pare PD, Hooper RO, Peretz DI,
Nelems JM, Schulzer M, and Hogg JC. The structure and
function of the pulmonary vasculature in mild chronic obstructive pulmonary disease: the effect of oxygen and exercise. Am
Rev Respir Dis 128: 702–707, 1983.
33. Zhuang FY, Fung YC, and Yen RT. Analysis of blood flow in
cat’s lung with detailed anatomical and elasticity data. J Appl
Physiol 55: 1341–1348, 1983.
91 • AUGUST 2001 •
www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.247 on June 16, 2017
16. Horsfield K. Morphometry of the small pulmonary arteries in
man. Circ Res 42: 593–597, 1978.
17. Horsfield K and Gordon WI. Morphometry of pulmonary veins
in man. Lung 159: 211–218, 1981.
18. Hyde RW. Circulation of lung neoplasms. In: The Bronchial
Circulation, edited by Butler J. New York: Marcel Dekker, 1992,
p. 551–578.
19. Jones R and Reid L. Vascular remodelling in clinical and
experimental pulmonary hypertensions. In: Pulmonary Vascular
Remodelling, edited by Bishop JE, Reeves JT, and Laurent GJ.
London: Portland, 1995, p. 47–116.
20. Majno G. Chronic inflammation: links with angiogenesis and
wound healing. Am J Pathol 153: 1035–1039, 1998.
21. Marshall BE and Marshall C. Pulmonary hypertension. In:
The Lung: Scientific Foundations (2nd ed.), edited by Crystal
RG, West JB, Barnes PJ, and Weibel ER. Philadelphia, PA:
Lippincott-Raven, 1997, p. 1581–1588.
22. Mattfeldt T, Frey H, and Rose C. Estimation of surface area
and length with the orientator. J Microsc 159: 301–317, 1990.
23. McCormack DG and Paterson NAM. Loss of hypoxic pulmonary vasoconstriction in chronic pneumonia is not mediated by
nitric oxide. Am J Physiol Heart Circ Physiol 265: H1523–
H1528, 1993.
24. Meyrick B and Reid L. Ultrastructural features of the distended pulmonary arteries of the normal rat. Anat Rec 193:
71–97, 1993.
25. Milne EN and Zerhouni EA. Blood supply of pulmonary metastases. J Thorac Imaging 2: 15–23, 1987.