Mouse strain dependence of lung tissue mechanics: Role of specific extracellular matrix composition

Respiratory Physiology & Neurobiology 152 (2006) 186–196

Mouse strain dependence of lung tissue mechanics: Role ofspecific extracellular matrix composition

Debora S. Faffe b, Elizabeth S. D’Alessandro a, Debora G. Xisto a,Mariana A. Antunes a, Pablo V. Romero c, Elnara M. Negri d,

Nilza R.D. Rodrigues e, Vera L. Capelozzi e,Walter A. Zin b, Patricia R.M. Rocco a,∗

a Laboratory of Pulmonary Investigation, Carlos Chagas Filho Biophysics Institute, Federal University of Rio de Janeiro,Centro de Ciencias da Saude, Ilha do Fundao, 21949-900 Rio de Janeiro, Brazil

b Laboratory of Respiration Physiology, Carlos Chagas Filho Biophysics Institute, Federal University of Rio de Janeiro,Centro de Ciencias da Saude, Ilha do Fundao, 21949-900 Rio de Janeiro, Brazil

c Laboratory of Experimental Pneumology, Ciutat Sanitaria de Bellvitge, IDIBELL, Barcelona, Spaind Laboratory of Cellular Biology (LIM59), University of Sao Paulo, Sao Paulo, Brazil

e Department of Pathology, University of Sao Paulo, Sao Paulo, Brazil

Accepted 13 August 2005

Abstract

This study analyses the differences between C57BL/10 and BALB/c mice in lung tissue micromechanical behaviour andwhether specific histological characteristics are related to the mechanical profile. C57BL/10 and BALB/c subpleural lung stripswere submitted to multisinusoidal deformation with frequencies ranging between 0.2 and 3.1 Hz. Tissue resistance (R), elastance(E), and hysteresivity (η) at each frequency were determined before and 30 s, 1, 2, and 3 min after acetylcholine (ACh) treatment.BALB/c mice showed higher E and R, at baseline, as well as greater amount of collagen and elastic fibres, and �-actin thanC57BL/10 mice. However, E, R, and η augmented with the same magnitude after ACh treatment in both strains. Baseline Rwas correlated with collagen fibre content and with the volume proportion of �-actin, while E was correlated with elastic andcollagen fibres, and �-actin contents. In conclusion, BALB/c and C57BL/10 mice present distinct tissue mechanical propertiesthat are accompanied by specific extracellular matrix composition and contractile structures.© 2005 Elsevier B.V. All rights reserved.

Keywords: Respiratory mechanics; Tissue mechanics; Mice; Strips; Extracellular matrix; Collagen fibres; Actin

∗ Corresponding author. Tel.: +55 21 2562 6557;fax: +55 21 2280 8193.

E-mail address: [email protected] (P.R.M. Rocco).

1. Introduction

It is generally accepted that asthmatic pulmonarydisorders result from inappropriate immune responses

1569-9048/$ – see front matter © 2005 Elsevier B.V. All rights reserved.doi:10.1016/j.resp.2005.08.008

D.S. Faffe et al. / Respiratory Physiology & Neurobiology 152 (2006) 186–196 187

to common environmental antigens in genetically sus-ceptible individuals (Ewart et al., 2000). In fact, sev-eral reports underline the role of genetic factors inhuman asthma (Doull et al., 1996; Barnes and Marsh,1998). Mice are widely used as in vivo asthma modelbecause of the availability of genetically character-ized inbred strains, and because this species allowsin vivo application of different immunological tools(Chiba et al., 1995; Brewer et al., 1999; Whitehead etal., 2003). Study of different mouse strains has beenwidely used to investigate the genetic mechanisms ofasthma-related phenotypes (Ewart et al., 1996, 2000;Drazen et al., 1999). Indeed, interstrain variations areobserved both in baseline conditions and after allergen-induced airway hyperresponsiveness (Brewer et al.,1999; Tankersley et al., 1999; Duguet et al., 2000; Heldand Uhlig, 2000; Takeda et al., 2001; Shinagawa andKojima, 2003).

There are evidences that asthmatic inflamma-tion extends to distal airways and lung parenchyma(Kuwano et al., 1993; Tulic et al., 2001; Tulic andHamid, 2003). Lung parenchyma strips are a goodproxy of the peripheral lung tissue and have beenextensively used to study contractile responses oflung periphery to several agonists in vitro (Ludwiget al., 1992; Fredberg et al., 1993). Two structuralcomponents are potentially involved in this phe-nomenon: the intrinsic mechanical changes of contrac-tile elements in the lung periphery (Helioui Raboudiet al., 1998) and the behaviour of collagen-elastinngnorapa

ttsclibac

with volume proportion of actin in the lung tissue ofeach strain.

2. Methods

2.1. Animal preparation

Seven C57BL/10 [BL (25–30 g)] and sevenBALB/c [BA (25–30 g)] mice were sedated (diazepam1 mg i.p.), anaesthetised [pentobarbital sodium(20 mg kg body weight−1 i.p.)], and heparine (1000 IU)was intravenously injected immediately before thesection of abdominal aorta and vena cava. The lungswere removed en bloc, and placed in a modifiedKrebs–Henseleith (K–H) solution [mM: 118.4 NaCl,4.7 KCl, 1.2 K3PO4, 25 NaHCO3, 2.5 CaCl2·H2O, 0.6MgSO4·H2O, and 11.1 glucose] at pH = 7.40 and 6 ◦C(Lopez-Aguilar and Romero, 1998; Faffe et al., 2001;Rocco et al., 2001, 2003a, 2003b).

2.2. Apparatus

Strips (2 mm × 2 mm × 10 mm) were cut from theperiphery of the left lung and suspended vertically ina K–H organ bath maintained at 37 ◦C, continuouslybubbled with a mixture of 95% O2–5% CO2. Lungstrips were weighed (W), and their unloaded restinglengths (L0) were determined with a calliper. Lungstrip volume was measured by simple densitometry,abδ

a

KaCBtBtwaNo(S

etwork (Mijailovich et al., 1993). It has been sug-ested that differences in airway-responsiveness phe-otype among different mouse strains is a propertyf the airways in all levels of the bronchial tree,ather than of the larger airways only (Duguet etl., 2000). Nevertheless, interstrain variations in lungarenchyma structural and functional characteristicsre unknown.

The aim of this study was to test the hypotheseshat different strains of mice could present differentissue mechanical profile, and that differences in tis-ue mechanics are potentially related to specific extra-ellular matrix components or contractile elements inung tissue. Tissue elastance, resistance and hysteresiv-ty were analysed during multisinosoidal oscillationsefore and after acetylcholine treatment in C57BL/10nd BALB/c mice. In addition, these parameters wereorrelated with collagen and elastic fibre contents, and

s: vol = �F/δ, where �F is the total change in forceefore and after strip immersion in K–H solution andis the mass density of K–H solution (Lopez-Aguilar

nd Romero, 1998; Romero et al., 2001).Parenchyma strips were suspended vertically in a

–H organ bath (30 ml internal volume) maintainedt 37 ◦C and continuously bubbled with 95% O2–5%O2 as previously described (Romero et al., 2001).riefly, one end of the strip was attached to a force

ransducer (LETICA TRI-110, Scientific Instruments,arcelona, Spain), whereas the other one was fas-

ened to a lever arm actuated by means of a modifiedoofer driven by the signal generated by a computer

nd analogue-to-digital converted (AT-MIO-16-E-10,ational Instruments, Austin, TX, USA). A sidearmf this rod was linked to a second force transducerLETICA TRI-110, Scientific Instruments, Barcelona,pain) by means of a silver spring of known Young’s

188 D.S. Faffe et al. / Respiratory Physiology & Neurobiology 152 (2006) 186–196

modulus, thus allowing the measurement of displace-ment. Neither amplitude dependence (<0.1% changein stiffness) nor phase changes with frequency weredetected in the range from 0.01 to 14 Hz. The hys-teresivity of the system (<0.003) was independent offrequency.

2.3. Preconditioning

Cross-sectional, unstressed area (A0) of the stripwas determined from volume and unstressed length,according to A0 = vol/L0. Basal force (FB) for a stress of0.1 N/cm2 was calculated as FB (N) = 0.1 (N/cm2). A0(cm2) and adjusted by vertical displacement of the forcetransducer as described before (Romero et al., 2001).The displacement signal was then set to zero. Oncebasal force and displacement signals were adjusted,the length between bindings (LB) was measured bymeans of a precision calliper. Instantaneous length dur-ing oscillation around LB was determined by addingthe value of LB to the measured value of displacementat any time. Instantaneous average cross-section area(Ai) was determined as Ai = Vs/Li (cm2). Instantaneousstress (σi) was calculated by dividing force (g) by Ai

(cm2). Strain was calculated as �∈ = (L − LB)/LB.Each parenchyma strip was preconditioned for

30 min by sinusoidal oscillation of the tissue (fre-quency = 0.5 Hz; amplitude large enough to reach afinal stress of 0.2 N/cm2). Thereafter, the amplitudewas adjusted to 5% L0 and the oscillation maintainedfwows

2

escscvpfm

of the composed signal (<0.04 cm peak to peak).Then, 3 baseline recordings (20-s duration) were done.Afterward, acetylcholine solution in K–H warmed at37 ◦C was added to the bath to reach successively bathconcentrations of 10−5, 10−3, and 10−1 M. Signalswere recorded at 30 s, 1, 2, and 3 min after each dose(20-s duration each). Both force and displacementsignals were preamplified, filtered at 30 Hz (902LPFFrequency Devices, Haverhill, MA, USA), analogue-to-digital converted (AT-MIO-16E-10, NationalInstruments Co., Marlboro, MA, USA) and sampledat a frequency of 150 Hz (Software LabWIEW 5.1,National Instruments Co., Austin, TX, USA).

2.5. Data analysis

Tissue elastance (E) and resistance (R), and hystere-sivity (η), were determined in the frequency domain (ω)as described before (Romero et al., 2001), according tothe following formulas:

Rω = R0ω−α (1)

Eω = E0ω−β (2)

η = ωR

E(3)

whereω is the angular frequency; R0 and E0 are, respec-tively, the values of tissue R and E at ω = 1 rad/s. α andβ are constants describing the frequency dependence

or another 30 min, or until a stable length-force loopas reached. After preconditioning, the strips werescillated at a frequency (f) = 1 Hz. The bath solutionas renewed regularly (every 20 min) with 37 ◦C K–H

olution.

.4. Experimental protocol

A total of 7 subpleural lung strips were studied forach group. After preconditioning, the samples wereubmitted to multifrequency forced oscillations. Theomputer-generated small-amplitude pseudorandomignal contained five noninteger discrete frequencyomponents of 0.2, 0.5, 1.1, 1.9, and 3.1 Hz. Indi-idual �∈ amplitudes were adjusted to equalize theower spectrum amplitude (±5% variation) at everyrequency. Individual phases were then adjusted toinimize peak-to-peak amplitude of the strain (�∈)

of R and E, respectively.Analysis and parameter estimation were performed

by means of specific software elaborated with Lab-VIEW 5.1 (National Instruments Co., Austin, TX,USA) as previously described (Romero et al., 2001).Briefly: the frequency response was computed fromthe time-domain strain (stimulus) and the time-domainstress (response) as:

Ψω = FFT (σ)

FFT ( ∈ )

where Ψ is the elastic or Young complex modulus(Ψ = σω/∈ω); FFT, fast Fourier transform; σ, the stress;and ∈, the strain. This result was transformed intosingle-sided magnitude (Ψ ) and phase (φ). E, R, andη were calculated according to:

Eω = Ψ cos (φ)

D.S. Faffe et al. / Respiratory Physiology & Neurobiology 152 (2006) 186–196 189

ηω = tan (φ)

Rω = Eω

ηω

ω

for the values of magnitude and phase correspondingto the relevant frequencies. Parameters β and E0 wereobtained by linear log–log correlation between ω andE� according to Eq. (2). Rυ (frequency-invariant com-ponent of R), R0, and α were obtained by using theLevenberg–Marquardt method to determine the leastsquares set of coefficients that best fit the set of inputdata points as expressed by [Rω = Rυ + R′

0ω−α].

2.6. Morphometric analysis

Lung tissue strips were fixed with 4% paraformalde-hyde. After fixation, the tissue was embedded in paraf-fin. Blocks were cut 4-�m-thick by a microtome. Formorphometric measurements, slices were stained withhaematoxylin-eosin and analysed with an integratingeyepiece with a coherent system made of a 100-pointgrid consisting of 50 lines of known length coupledto a light microscope (Axioplan, Zeiss, Oberkochen,Germany). Sections were examined at 400× mag-nification, and the fractional areas of alveolar wall(AW), blood-vessel wall (BVW), and bronchial wall(BW) were determined by the point-counting technique(Weibel, 1990). All points falling on these compo-nents were counted and divided by the total numberof points. This analysis was performed in 10 random,nwlcse

ofisoltosstt

ware (Bioscan-Optimas 5:1, Bioscan, Edmond, WA,USA). The images were generated by a microscope(Axioplan, Zeiss, Oberkochen, Germany) connected toa camera (Trinitron CCD, Sony, Tokyo, Japan) and fedinto a computer through a frame grabber (Oculus TCX,Coreco, St. Laurent, Que., Canada) for off-line pro-cessing. The thresholds for fibres of the collagenousand elastic systems were established after enhancingthe contrast up to a point at which the fibres wereeasily identified as either black (elastic) or birefrin-gent (collagen) bands. The area occupied by fibreswas determined by digital densitometric recognition.Bronchi and blood vessels were carefully avoided dur-ing the measurements. To avoid any bias due to septumoedema or alveolar collapse, the areas occupied by theelastic and collagen fibres, measured in each alveolarseptum, were divided by the length of the correspond-ing septum. The results were expressed as the num-ber of elastic and collagen fibres per unit of septumlength.

2.7. Smooth-muscle-specific actin evaluation

Immunohistochemical staining was performed inslides 5-�m-thick of lung tissue strips fixed with 4%paraformaldehyde, using monoclonal antibody to �-smooth muscle actin (Dako, Carpenteria, CA, USA)at a 1:500 dilution. Sections were then rinsed withTris-buffered saline and incubated sequentially withbiotinylated rabbit antimouse IgG (Dako Corp., Cam-bvpbdctmebwt1sabts

on-overlapping fields in each strip. BVW and BWere counted when a point fell on the endothelial

ayer, the epithelial layer, the smooth muscle, or asso-iated connective tissue. Points falling on alveolar airpaces, blood-vessel lumen, and bronchial lumen werexcluded.

Tissue slices also underwent specific staining meth-ds to characterise the collagenous and elastic systembres in the alveolar septa. For collagen, the tissue wastained in a solution of Sirius red dissolved in aque-us saturated picric acid and observed under polarizedight microscopy (Montes, 1996). For elastic fibres,he Weigert’s resorcin fuchsin method modified withxidation was used (Fullmer et al., 1974). In eachtrip, 20 different microscopic fields were randomlyelected to quantify collagen and elastic fibres. Quan-ification (200× magnification) was carried out withhe aid of a digital analysis system, using specific soft-

ridge, UK) at a dilution of 1:400, followed by strepta-idin combined in vitro with biotinylated horseradisheroxidase at a dilution of 1:1000 (Dako, Cam-ridge, UK). The reaction product was developed usingiaminobenzidine tetrahydrochloride. Sections wereounterstained with hematoxylin for 1 min, dehydratedhrough graded alcohols, and mounted in resinous

ountant. Known positive controls were included withach run, and negative controls had the primary anti-ody omitted (Dolhnikoff et al., 1998). The analysisas performed on the slides stained for actin applying

he point-counting technique (Weibel, 1990). Using a21-point grid, we calculated the volume proportion ofmooth-muscle-specific actin in terminal bronchiolesnd alveolar ducts as the relation between the num-er of points falling on actin-stained and non-stainedissue. Measurements were done at 400× in eachlide.

190 D.S. Faffe et al. / Respiratory Physiology & Neurobiology 152 (2006) 186–196

2.8. Statistical analysis

SigmaStat 2.0 statistical software package (JandelCorporation, CA, USA) was used. Differences betweenthe two groups were assessed by Student’s t-test. One-way repeated measures ANOVA was applied to test thesignificance of variations of measured and calculatedparameters, during basal conditions and diverse acetyl-choline challenges. Correlation between mechanicaland histological data was determined by Spearmancorrelation test. A P-value <0.05 was considered sig-nificant.

3. Results

The effects of different frequencies on tissue elas-tance (E), resistance (R), and hysteresivity (η) areshown in Fig. 1. In both strains, R had a negative and Ea positive dependence of frequency, while η remainedunchanged. Elastance and resistance were higher inBALB/c than in C57BL/10 mice at all frequencies stud-ied (P < 0.05).

Tissue mechanics were measured at 30 s, 1, 2, and3 min after acetylcholine treatment, and the time pointpresenting the highest value of R was used for subse-quent analysis for all mechanical parameters. R, E, andη increased in both groups after 10−1 M acetylcholinechallenge (Fig. 1) (P < 0.05). Lower doses of acetyl-choline (10−3 and 10−5 M) had no significant effect ontptm

pvShTcBdPcyph

Fig. 1. Tissue elastance, resistance, and hysteresivity (η) inC57BL/10 and BALB/c mice at different frequencies (0.2, 0.5,1.1, 1.9, and 3.1 Hz) and after 10−1 M acetylcholine (ACh) chal-lenge. Values are mean ± S.E.M. of seven animals in each group.(#) P < 0.05 baseline values in comparison to the values after AChtreatment; (&) P < 0.05 baseline values between the two strains; (*)P < 0.05 values after ACh challenge between the two strains.

issue mechanics in both strains (data not shown). Theercentage increase in E, R, and η after acetylcholinereatment was similar between BALB/c and C57BL/10

ice in all frequencies.Results of the parametric analysis of the model are

resented in Table 1. Differences between basal meanalues before acetylcholine challenge were tested bytudent’s t-test, baseline E0 and β were significantlyigher in BALB/c than in C57BL/10 mice (P < 0.05).he frequency-dependent (R′

0) and -independent (Rυ)omponents of tissue R also tended to be higher inALB/c than C57BL/10 although these differencesid not achieved statistic significance (P = 0.07 and= 0.05, respectively). ANOVA showed significant

hanges for R′0, E0, β, and α + β (constant-phase anal-

sis) after acetylcholine (10−1 M) stimulation com-ared to control in both strains. Nevertheless, even theighest dose of ACh failed to modify the frequency-

D.S. Faffe et al. / Respiratory Physiology & Neurobiology 152 (2006) 186–196 191

Table 1Model parameters during acetylcholine stimulation of lung parenchyma strips in C57BL/10 and BALB/c mice

Basal ACh (10−5 M) ACh (10−3 M) ACh (10−1 M) P

C57BL/10R′

0 (102 s/m2) 6.75 ± 1.08 6.35 ± 1.33 7.71 ± 1.77 7.68 ± 1.17 <0.001α −1.005 ± 0.141 −0.978 ± 0.090 −1.049 ± 0.084 −1.052 ± 0.131 0.678E0 (104 N/m2) 1.25 ± 0.09 1.24 ± 0.10 1.28 ± 0.10 1.31 ± 0.10 0.001β 0.029 ± 0.004 0.028 ± 0.004 0.029 ± 0.002 0.031 ± 0.002 0.006Rυ (102 N s/m2) 0.35 ± 0.10 0.35 ± 0.08 0.34 ± 0.06 0.34 ± 0.09 0.081α + β 1.034 ± 0.140 1.006 ± 0.091 1.070 ± 0.085 1.083 ± 0.130 0.006

BALB/cR′

0 (102 N s/m2) 7.93 ± 1.10 8.81 ± 1.07 10.13 ± 1.74 10.80 ± 2.45 <0.001α −0.947 ± 0.108 −1.000 ± 0.093 −1.103 ± 0.081 −1.111 ± 0.139 0.073E0 (104 N/m2) 1.45 ± 0.15 1.46 ± 0.14 1.49 ± 0.15 1.53 ± 0.15 0.002β 0.034 ± 0.004 0.032 ± 0.004 0.031 ± 0.004 0.034 ± 0.003 0.011Rυ (102 N s/m2) 0.48 ± 0.13 0.49 ± 0.17 0.39 ± 0.16 0.38 ± 0.15 0.084α + β 0.981 ± 0.108 1.032 ± 0.092 1.135 ± 0.079 1.145 ± 0.139 0.012

Values are mean ± S.D. of 7 strips in each group. R′0, tissue resistance (R) component hyperbolically decreasing with frequency at ω = 1 rad/s;

E0, value of tissue elastance (E) at ω = 1 rad/s; Rυ , frequency-invariant component of R; α and β, constant describing the frequency-dependenceof R and E, respectively. P-values for ACh (10−1 M) in comparison to basal.

independent component of tissue resistance (Rυ), aswell as α (Table 1). Mean values of R′

0 and E0 after10−1 M ACh were significantly higher in BALB/c thanC57BL/10 mice (P < 0.05).

Table 2 shows the results of morphometrical anal-ysis. Although both groups showed similar anatomiccomposition (amount of BW, BVW, and AW), theamount of collagen and elastic fibres and the volumeproportion of �-actin in the lung periphery were higherin BALB/c than in C57BL/10 mice (Table 3, Fig. 2).

In order to determine whether interstrain differencesin tissue mechanics are influenced by the differencesobserved in lung parenchyma components, the corre-lation between mechanical parameters at 1.1 Hz beforeacetylcholine treatment and the morphometric data wasanalysed (Fig. 3). E and R were significantly correlatedwith the amount of collagen fibres and with the volumeproportion of �-actin, while E was also correlated withthe amount of elastic fibres.

Table 2Morphometrical parameters in different mouse strains

AW (%) BVW (%) BW (%)

C57BL/10 97.42 ± 2.00 0.67 ± 0.67 1.91 ± 0.93BALB/c 98.05 ± 2.02 0.37 ± 0.59 1.57 ± 1.49

Values are mean ± S.D. of 7 strips in each group (10 random, non-coincident microscopic fields were analysed in each strip) (AW,alveolar wall; BVW, blood vessel wall; BW, bronchial wall).

Table 3Amounts of elastic and collagen fibres in alveolar walls and vol-ume proportion of �-smooth muscle actin in parenchymal strips ofdifferent strains of mice

Collagen fibre(�m2/�m)

Elastic fibre(�m2/�m)

�-actin (%)

C57BL/10 0.015 ± 0.003 0.27 ± 0.01 13.50 ± 3.40BALB/c 0.030 ± 0.006* 0.31 ± 0.002* 30.83 ± 5.70*

Values are mean ± S.D. of 7 animals in each group (10 random, non-coincident microscopic fields were analysed in each lung).

* Values significantly different from C57BL/10 (P < 0.05).

4. Discussion

This study shows that tissue elastance and resis-tance are higher in naıve BALB/c than in C57BL/10mice (Fig. 1). Both strains, however, presented the samemagnitude of changes in all tissue mechanical param-eters after acetylcholine challenge. BALB/c mice havehigher amounts of collagen and elastic fibres, as wellas higher volume proportion of �-actin in the lungparenchyma than C57BL/10 mice (Table 3, Fig. 2).These differences in extracellular matrix compositionmay be related to the differences observed in tissuemechanics between the two strains, since a correlationbetween morphological and functional data was found(Fig. 3).

Study of different mouse strains has been widelyused to investigate the genetic mechanisms of asthma-

192 D.S. Faffe et al. / Respiratory Physiology & Neurobiology 152 (2006) 186–196

Fig. 2. Immunohistochemical staining for smooth-muscle-specificactin in C57BL/10 (A); and BALB/c (B) mice. Note positive stainingfor actin in terminal bronchiole (TB) and alveolar duct (AD) (arrows)(arrowhead: arteriole; scale bars = 100 �m).

related phenotypes (Ewart et al., 1996; Drazen et al.,1999; Ewart et al., 2000). Murine models of allergicasthma avoid factors such as variability in clinicalphenotype, uncontrolled environmental influences,and genetic heterogeneity within studied populationusually observed in human asthma (Ewart et al., 2000).Indeed, interstrain differences have been reportedboth in allergen-induced airway hyperresponsivenessmodels and in naıve mices (Brewer et al., 1999;Drazen et al., 1999; Duguet et al., 2000; Ewart et al.,2000; Held and Uhlig, 2000).

BALB/c and C57BL/10 mice were selected accord-ing to their usage in lung physiological and immuno-logical investigations (Held and Uhlig, 2000). BALB/cis frequently used in laboratory animal research, and in

immunologic studies, while C57BL/10 mice serve asthe background strain of many genetically engineeredmice. BALB/c strain displays significant increase inairway responsiveness, eosinophilia, and IgE produc-tion after allergen sensitization and challenge, whileC57BL/10 strain represents one of the least responsivestrains in the development of airway hyperresponsive-ness (Chiba et al., 1995; Brewer et al., 1999; Drazen etal., 1999; Held and Uhlig, 2000; Takeda et al., 2001;Whitehead et al., 2003).

Although previous evidences suggest that differ-ences in the airway-responsiveness phenotype amongmouse strains occur across the entire airway tree(Duguet et al., 2000), interstrain variations in lungparenchyma structural and functional characteristicsare poorly understood. The present study focusedon determining the differences between BALB/c andC57BL/10 mice regarding lung parenchyma.

Physically, lung parenchyma can be simplified as aviscoelastic connective matrix connected to a contrac-tile system that modulates its mechanical properties.It is currently accepted that the connective tissue fibrenetwork dominates parenchymal mechanics, togetherwith a less significant role played by interstitial cells(Fredberg et al., 1993; Yuan et al., 1997). However, theeffect of tissue contraction on the rheological proper-ties of lung tissue is not completely understood.

In our study, η did not show significant changewith frequency at baseline. After ACh challenge η

remained invariant between 1.9 and 3.1 Hz, but tendedtTifssoscnwnIwpvAm

o increase at the lowest frequencies (0.2 and 0.5 Hz).his behaviour was similar to that previously observed

n histamine-challenged guinea-pig strips under multi-requency inputs (Romero et al., 2001), while othertudy using the multifrequency approach showed amall but not constant positive frequency dependencef η (Yuan et al., 1997). It is reported that η can showome frequency-dependent behaviour when the vis-oelastic model for lung parenchyma behaviour doesot meet the constant-phase requirement α + β = 1, orhen the frequency-invariant element of resistance isot negligible (Suki et al., 1994; Romero et al., 2001).n our samples, no significant change in Rυ after AChas observed, but when the reliability of the constant-hase hypothesis was tested by comparing α and 1 − β

alues, significance was achieved at 10−3 and 10−1 MCh (P < 0.001 and P = 0.02, respectively) in BALB/cice.

D.S. Faffe et al. / Respiratory Physiology & Neurobiology 152 (2006) 186–196 193

Fig. 3. Correlations between tissue mechanical data at 1.1 Hz before acetylcholine challenge (baseline) and morphological data obtained bySpearman’s correlation test in C57BL/10 (�) and BALB/c (©) (7 strips/group). Collagen and elastic fibres are expressed as the amountof each fibre per unit septum length (�m2/�m). Actin (%) expresses the volume proportion of �-smooth muscle actin in lung strips byimmunohistochemical staining.

It was postulated that two mechanisms contributeto the constant-phase tissue viscoelasticity: thestructural disposition of fibers and their instantaneousconfiguration during motion (Suki et al., 1994),and the interaction between fibers in close proximity(Mijailovich et al., 1993). Both mechanisms contributeto maintain the viscoelastic properties of lung tissueas long as the architectural microstructure is not mod-ified, as during passive stretch. Larger constrictions,however, are expected to rearrange the connective

matrix, changing the microstructure of the lung, andso challenging the constant-phase behaviour (Romeroet al., 2001). Our results showed that C57BL/10tissue mechanical response to acetylcholine obeysthe constant-phase hypothesis (Hantos et al., 1992),whereas for tissue properties of BALB/c mice at highdegree of constriction the constant-phase hypothesiswas not satisfied. Since BALB/c mice presented higheramount of collagen and elastic fibres, and higher con-tent of �-actin in lung tissue than C57BL/10, these

194 D.S. Faffe et al. / Respiratory Physiology & Neurobiology 152 (2006) 186–196

specific structural differences might have potentiallycontributed to the observed behaviour after AChchallenge. Our findings are in agreement with aprevious observation that histamine-induced pneumo-constriction in guinea pigs rejects the constant-phasehypothesis at 10−5 and 10−3 M (Romero et al., 2001).At these high contractions, α was significantly differ-ent from 1 − β, there was a frequency-independentcomponent of resistance significantly different fromnil, and hysteresivity developed a frequency-dependentbehaviour. The authors attributed the failure of theconstant-phase hypothesis to muscular contraction orextracellular matrix rearrangement. Although a recentstudy reported oscillatory tissue strip data that fit theconstant-phase (Fust et al., 2004), it is noteworthy thatstrips were not submitted to contractile stimulation.Indeed, Romero et al. did not reject the constant-phasehypothesis either during baseline state or duringpassive stretching of the sample, but only at the higherdegrees of constrictor challenge (Romero et al., 2001).

The anatomic elements that potentially determinetissue viscoplastoelastic behaviour include the networkof stress-bearing collagen and elastic fibres, proteo-glycans and glycosaminoglycans and the contractileelements present in parenchymal tissues (Fredberg andStamenovic, 1989; Fredberg et al., 1993; Mijailovichet al., 1993; Cavalcante et al., 2005). We demonstratedthat BALB/c mice have a higher content of collagenand elastic fibres in comparison to C57BL/10 mice.The amount of collagen fibres was positively corre-lpststtocdifTtubcs

vation of specific differences in lung periphery. On theother hand, we cannot rule out a potential effect of inter-strain differences in the amount and/or compositionof proteoglycans. Previous studies report that swellingof charged proteoglycans influences the macroscopicmechanical properties of lung tissue strips. Although itappears that the network organization of collagen andelastic fibres is the dominant factor in elastic behaviourof lung tissue, interaction of the fibres with proteogly-cans can affect fibre network stability, thus contributingto the tissue mechanical profile (Cavalcante et al., 2005;Al Jamal et al., 2001).

Although there was no significant difference in thevolume proportion of BW, BVW, and AW among thestrips, the present study showed that the volume propor-tion of �-actin is higher in BALB/c lung parenchymain comparison to C57BL/10, yielding to higher val-ues of tissue resistance and elastance in the former(Fig. 1). Thus, the differences in the mechanical proper-ties between the two strains can probably be attributedto differences in extracellular matrix composition andvolume proportion of �-actin.

Tissue elastance, resistance and hysteresivityincreased with the stimulation of lung tissue withacetylcholine. This response is in accordance withother studies (Fredberg et al., 1993; Salerno et al., 1995,2004) where constrictor response was observed whenlung parenchymal strips were challenged with smoothmuscle agonists. Small airways, small blood vessels,and interstitial contractile cells may be responsible fortttcchpsntwmotdcfip

ated with R and E, while elastic fibre content wasositively correlated with tissue elastance. These datauggest that collagen and elastic fibres contributed tohe differences in stiffness and in viscosity of lung tis-ue between the two strains. It is noteworthy, however,hat not only the absolute amount of fibres is impor-ant in affecting mechanical behaviour, but rather therganization and/or the interaction of these fibres. Inontrast to our findings, Shinagawa and Kojima (2003)id not report baseline differences in levels of hydrox-proline (total collagen) in lung tissue when studyingour mouse strains exposed to intranasal ovalbumin.his discrepancy could be explained by the fact that

otal collagen was analysed in the whole lung, thus,nder the influence of major bronchial, airway andlood vessels. In our study, collagen and elastic fibreontents were measured in isolated lung parenchymatrips that could have potentially allowed the obser-

he contractile response of lung periphery. In this con-ext, Fredberg et al. (1993) showed that changes in hys-eresivity are directly associated with the cross-bridgeycling rate and hence with the metabolic state of theells, while in the absence of stimulation the value ofysteresivity is attributable to the passive mechanicalroperties of connective tissue. In our study, η wasimilar in both groups, and increased in the same mag-itude after acetylcholine treatment. Thus, althoughhe amount of elastic and collagen fibres and �-actinere higher in BALB/c in comparison to C57BL/10ice, the mechanical interaction and the arrangement

f fibres and contractile cells in the interstitium yieldedo a similar η in both strains under baseline con-itions. On the other hand, the addition of acetyl-holine in the bath stretches the surrounding connectivebres during contraction changing their mechanicalroperties.

D.S. Faffe et al. / Respiratory Physiology & Neurobiology 152 (2006) 186–196 195

Although differences in baseline lung tissue prop-erties between BALB/c and C57BL/10 mice wereobserved, there was no effect of the genetic back-ground (mouse strain) on their response to acetyl-choline. There is a lot of evidence in the literature thatinbred mouse strains exhibit significant genetic vari-ability in their susceptibilities to develop asthma-likephenotype (Brewer et al., 1999; Drazen et al., 1999;Shinagawa and Kojima, 2003; Whitehead et al., 2003).Nevertheless, it has been reported that noninflamma-tory and allergen-induced airway responsiveness areuniquely regulated, suggesting that airway contractionin these two situations are controlled by different genes(Ewart et al., 2000). Based on the aforementioned, oneshould not necessarily expect similar lung peripheryresponse in BALB/c and C57BL/10 strains after aller-gen stimulation.

In conclusion, our findings showed that BALB/cand C57BL/10 mice present distinct tissue mechanicalproperties that were related to specific extra-cellularfibre composition and amount of contractile struc-tures in their lung parenchyma. This study suggeststhat genetic background determines basal lung tissuefunction and structural differences between these twostrains.

Acknowledgements

We would like to express our gratitude to Mr. Anto-nd

lTiFdpR

R

B

B

Cavalcante, F.S., Ito, S., Brewer, K., Sakai, H., Alencar, A.M.,Almeida, M.P., Andrade Jr., J.S., Majumdar, A., Ingenito, E.P.,Suki, B., 2005. Mechanical interactions between collagen andproteoglycans: implications for the stability of lung tissue. J.Appl. Physiol. 98, 672–679.

Chiba, Y., Yanagisawa, R., Sagai, M., 1995. Strain and route dif-ferences in airway responsiveness to acetylcholine in mice. Res.Commun. Mol. Pathol. Pharmacol. 90, 169–172.

Dolhnikoff, M., Morin, J., Ludwig, M.S., 1998. Human lungparenchyma responds to contractile stimulation. Am. J. Respir.Crit. Care Med. 158, 1607–1612.

Doull, I.J., Lawrence, S., Watson, N.E., 1996. Allelic associationof gene markers on chromossomes 5q and 11q with atopy andbronchial hyperresponsiveness. Am. J. Respir. Crit. Care Med.153, 1280–1284.

Drazen, J.M., Finn, P.W., De Sanctis, G.T., 1999. Mouse models ofairway responsiveness: physiological basis of observed outcomesand analysis of selected examples using these outcome indicators.Annu. Rev. Physiol. 61, 593–625.

Duguet, A., Biyah, K., Minshall, E., Gomes, R., Wang, C.,Taoudi-Benchekroun, M., Bates, J.H.T., Eidelman, D.H., 2000.Bronchial responsiveness among inbred mouse strains. Role ofairway smooth-muscle shortening velocity. Am. J. Respir. Crit.Care Med. 161, 839–848.

Ewart, S.L., Mitzner, W., DiSilvestre, D.A., Meyers, D.A., Levitt,R.C., 1996. Airway hyperresponsiveness to acetylcholine: segre-gation analysis and evidence for linkage to murine chromossome6. Am. J. Respir. Cell Mol. Biol. 14, 4887–4895.

Ewart, S.L., Kuperman, D., Schadt, E., Tankersley, C., Grupe, A.,Shubitowski, D.M., Peltz, G., Wills-Karp, M., 2000. Quantitativetrait loci controlling allergen-induced airway hyperresponsive-ness in inbred mice. Am. J. Respir. Cell Mol. Biol. 23, 537–545.

Faffe, D.S., Silva, G.H., Kurtz, P.M.P., Negri, E.M., Capelozzi, V.L.,Rocco, P.R.M., Zin, W.A., 2001. Lung tissue mechanics and

F

F

F

F

H

H

H

io Carlos de Souza Quaresma and Veronica Cristinaos Santos for their skillful technical assistance.

This study was supported by The Centres of Excel-ence Program (PRONEX-MCT and MCT/FAPERJ),he Brazilian Council for Scientific and Technolog-

cal Development (MCT/CNPq), The Carlos Chagasilho Rio de Janeiro State Research Supporting Foun-ation (FAPERJ), The Sao Paulo State Research Sup-ort Foundation (FAPESP), and Red Respira (ISCiiiTIC 03/011) Spain.

eferences

arnes, K.C., Marsh, D.G., 1998. The genetics and complexity ofallergy and asthma. Immunol. Today 19, 325–332.

rewer, J.P., Kisselgof, A.B., Martin, T.R., 1999. Genetic variabilityin pulmonary physiological, cellular, and antibody responses toantigen in mice. Am. J. Respir. Crit. Care Med. 160, 1150–1156.

extracellular matrix composition in a murine model of silicosis.J. Appl. Physiol. 90, 1400–1406.

redberg, J.J., Bunk, D., Ingenito, E.P., Shore, S.A., 1993. Tissueresistance and the contractile state of lung parenchyma. J. Appl.Physiol. 74, 1387–1397.

redberg, J.J., Stamenovic, D., 1989. On the imperfect elasticity oflung tissue. J. Appl. Physiol. 67, 2408–2414.

ullmer, H.M., Sheetz, J.H., Narkates, A.J., 1974. Oxytalan connec-tive tissue fibers: a review. J. Oral Pathol. 3, 291–316.

ust, A., Bates, J.H., Ludwig, M.S., 2004. Mechanical properties ofmouse distal lung: in vivo versus in vitro comparison. Respir.Physiol. Neurobiol. 143, 77–86.

antos, Z., Daroczy, B., Suki, B., Nagy, S., Fredberg, J.J., 1992.Input impedance and peripheral inhomogeneity of dog lungs. J.Appl. Physiol. 72, 168–178.

elioui Raboudi, S., Miller, B., Butler, J.P., Shore, S.A., Fredberg,J.J., 1998. Dynamically determined contractile states of airwaysmooth muscle. Am. J. Respir. Crit. Care Med. 158, S176–S178.

eld, H., Uhlig, S., 2000. Basal lung mechanics and airway andpulmonary vascular responsiveness in different inbred mousestrains. J. Appl. Physiol. 88, 2192–2198.

196 D.S. Faffe et al. / Respiratory Physiology & Neurobiology 152 (2006) 186–196

Al Jamal, R., Roughley, P.J., Ludwig, M.S., 2001. Effect of gly-cosaminoglycan degradation on lung tissue viscoelasticity. Am.J. Physiol. Lung Cell. Mol. Physiol. 280, L306–L315.

Kuwano, K., Bosken, C.H., Pare, P.D., Bai, T.R., Wiggs, B.R.,Hogg, J.C., 1993. Small airways dimensions in asthma and inchronic obstructive pulmonary disease. Am. Rev. Respir. Dis.148, 1220–1225.

Lopez-Aguilar, J., Romero, P.V., 1998. Effect of elastase pretreat-ment on rat lung strip induced constriction. Respir. Physiol. 113,239–246.

Ludwig, M.S., Robatto, F.M., Simard, S., Stamenovic, D., Fredberg,J.J., 1992. Lung tissue resistance during contractile stimula-tion: structural damping decomposition. J. Appl. Physiol. 72,1332–1337.

Mijailovich, S.M., Stamenovic, D., Fredberg, J.J., 1993. Towarda kinetic theory of connective tissue micromechanics. J. Appl.Physiol. 74, 665–681.

Montes, G.S., 1996. Structural biology of the fibers of the collage-nous and elastic system. Cell. Biol. Int. 20, 15–27.

Rocco, P.R.M., Negri, E.M., Kurtz, P.M., Vasconcellos, F.P., Silva,G.H., Capelozzi, V.L., Romero, P.V., Zin, W.A., 2001. Lung tis-sue mechanics and extracellular matrix in acute lung injury. Am.J. Respir. Crit. Care Med. 164, 1067–1071.

Rocco, P.R.M., Souza, A.B., Faffe, D.S., Passaro, C.P., Santos, F.B.,Negri, E.M., Lima, J.G.M., Contador, R.S., Capelozzi, V.L., Zin,W.A., 2003a. Effect of corticosteroid on lung parenchyma remod-elling at an early phase of acute lung injury. Am. J. Respir. Crit.Care Med. 168, 677–684.

Rocco, P.R.M., Momesso, D.P., Figueira, R.C., Ferreira, H.C.,Cadete, R.A., Legora-Machado, A., Koatz, V.L., Lima, L.M.,Barreiro, E.J., Zin, W.A., 2003b. Therapeutic potential of a newphosphodiesterase inhibitor in acute lung injury. Eur. Respir. J.22, 20–27.

Romero, P.V., Zin, W.A., Lopez-Aguilar, J., 2001. Frequency char-acteristics of lung tissue during passive stretch and induced

Salerno, F.G., Dallaire, M., Ludwig, M.S., 1995. Does the anatomicmakeup of parenchymal lung strips affect oscillatory mechan-ics during induced constriction? J. Appl. Physiol. 79, 66–72.

Salerno, F.G., Fust, A., Ludwig, M.S., 2004. Stretch-induced changesin constricted lung parenchymal strips: role of extracellularmatrix. Eur. Respir. J. 23, 193–198.

Shinagawa, K., Kojima, M., 2003. Mouse model of airway remod-elling. Strain differences. Am. J. Respir. Crit. Care Med. 168,959–967.

Suki, B., Barabasi, A.-L., Lutchen, K.R., 1994. Lung tissue viscoelas-ticity: a mathematical framework and its molecular basis. J. Appl.Physiol. 76, 2749–2759.

Takeda, K., Haczku, A., Lee, J.J., Irvin, C.G., Gelfand, E.W., 2001.Strain dependence of airway hyperresponsiveness reflects differ-ences in eosinophil localization in the lung. Am. J. Physiol. LungCell Mol. Physiol. 281, L394–L402.

Tankersley, C.G., Rabold, R., Mitzer, W., 1999. Differential lungmechanics are genetically determined in inbred murine strains.J. Appl. Physiol. 86, 1764–1769.

Tulic, M.K., Christodoulopoulos, P., Hamid, Q., 2001. Small airwayinflammation in asthma. Respir. Res. 2, 333–339.

Tulic, M.K., Hamid, Q., 2003. Contribution of the distal lung to thepathologic and physiologic changes in asthma: potential thera-peutic target. Chest 123 (Suppl.), 348–355.

Weibel, E.R., 1990. Morphometry: stereological theory and practicalmethods. In: Gil, J. (Ed.), Models of Lung Disease-Microscopyand Structural Methods. Marcel Decker, New York, pp. 199–247.

Whitehead, G.S., Walker, J.K.L., Berman, K.G., Foster, W.M.,Schwartz, D.V., 2003. Allergen-induced airway disease is mousestrain dependent. Am. J. Physiol. Lung Cell Mol. Physiol. 285,32–42.

Yuan, H., Ingenito, E.P., Suki, B., 1997. Dynamic properties of lungparenchyma: mechanical contributions of fiber network and inter-

pneumoconstriction. J. Appl. Physiol. 91, 882–890.