Cilia-driven particle and fluid transport over mucus-free mice tracheae

(1)

Cilia-driven particle and ﬂuid transport over mucus-free mice tracheae

J. Hussong

f,n

, R. Lindken

b

, P. Faulhammer

c

, K. Noreikat

d

, K.V. Sharp

e

, W. Kummer

c

,

J. Westerweel

a

Laboratory for Aero- and Hydrodynamics, Leeghwaterstraat 21, 2628CA Delft, The Netherlands

b

Zentrum f¨ur Brennstoffzellen Technik, ZBT GmbH, Abteilung Mikrosysteme und Str¨omungsmechanik, Carl-Benz-Strasse 201, 47057 Duisburg, Germany

c

Institut f¨ur Anatomie und Zellbiologie, Justus-Liebig-Universit¨at Giessen, Germany

d

Department of Anesthesiology and Intensive Care Medicine, University of Leipzig, 04103 Leipzig, Germany

e

School of Mechanical, Industrial, & Manufacturing Engineering, 204 Rogers Hall, Oregon State University, Corvallis, OR 97331-6001, USA

f_{Chair of Hydraulic Fluid Machinery, Institute for Thermo- and Fluiddynamics, Universit¨}_{atsstraße 150, 44801 Bochum, Germany}

a r t i c l e

i n f o

Article history: Accepted 9 August 2012 Keywords: ASL transport PCL transport Mouse trachea

m

PIV

a b s t r a c t

To date, there is only a fragmentary understanding of the fundamental mechanisms of airway mucociliary transport. Application of the latest measurement techniques can aid in deciphering the complex interplay between ciliary beat and airway surface liquid (ASL) transport. In the present study, direct, quasi-simultaneous measurements of the cilia-induced ﬂuid and bead transport were performed to gain a better insight into both transport mechanisms. In this study cilia-induced periciliary liquid (PCL) transport is measured by means of micro Particle Image Velocimetry (

m

PIV) with neutrally buoyant tracers. Particle Tracking Velocimetry (PTV) with heavier polystyrene-ferrite beads is performed to simulate particle transport. Contrary to recent literature, in which the presence of mucus was deemed necessary to maintain periciliary liquid (PCL) transport, effective particle and ﬂuid transport was measured in our experiments in the absence of mucus. In response to muscarine or ATP stimulation, maximum ﬂuid transport rates of 250

m

m=s at 15

m

m distance to the tracheal epithelia were measured while bead transport rates over the epithelia surfaces reached 200

m

m=s. We estimated that the mean bead transport is dominated by viscous drag compared to inertial fluid forces. Furthermore, mean bead transport velocities appear to be two orders of magnitude larger compared to bead sedimentation velocities. Therefore, beads are expected to closely follow the mean PCL flow in non-ciliated epithelium regions. Based on our results, we have shown that PCL transport can be directly driven by the cilia beat and that the PCL motion may be capable of driving bead transport by fluid drag. &2012 Published by Elsevier Ltd.

1. Introduction

Mucociliary clearance is a primary innate defense mechanism

that serves to eliminate inhaled particles and bacteria from the

conducting airways of the respiratory tract. Its structural basis

consists of the respiratory epithelium and the airway surface liquid

(ASL). It is composed of a tangled polymer network of large and

extensively glycosylated macromolecules (mucins) that is referred

to as mucus, and an underlying low-viscosity periciliary liquid (PCL)

(

Knowles and Boucher, 2002

).

Fig. 1

a shows a sketch of a ciliated

epithelium with ASL. The driving force of ASL transport is generated

by beating cilia. Airway cilia are cylindrical cell extensions with a

typical length of 510

m

m and a typical thickness of 0:25

m

m that

actively beat with a frequency of 15–30 Hz by ATP-dependent

tubulin–dynein interaction (

Crystal et al., 1997

). Ciliated airway

epithelial cells carry approximately 160–200 of such cilia that

co-ordinately beat towards the larynx. Basal, secretory and ciliated

cells form the epithelium that line the inside of the airway. A ciliated

epithelium of a mouse trachea can be seen in

Fig. 1

c. Disorders in

ASL production or transport, such as impaired ciliary beat in

immotile cilia syndromes and imbalance of ASL composition in

cystic ﬁbrosis, have severe impact on mucociliary clearance, and can

lead to serious, even life-shortening diseases (

Donaldson and

Boucher, 2007

;

Mall, 2008

;

M ¨oller et al., 2006

;

Mossberg et al.,

1978

). Despite extensive research, the mechanical coupling of ciliary

beat to effective ASL transport is only fragmentarily understood and

a matter of ongoing discussion (

Braiman and Priel, 2008

;

Donaldson

et al., 2006

;

Fahy and Dickey, 2010

;

Francis et al., 2009

;

Klein et al.,

2009

;

Knowles and Boucher, 2002

;

K ¨onig et al., 2009

;

Mall, 2008

;

Sleigh et al., 1988

). According to the currently accepted concept, the

ciliary tips touch the mucus layer during the power stroke, but not

so during the recovery stroke. As a consequence, they would act on

the lower surface of the mucus sheet to move mucus

unidirection-ally, and periciliary ﬂuid transport would be initiated by the

frictional interaction with the mucus layer (

Knowles and Boucher,

2002

). The model originates from cell culture experiments of

Contents lists available at

SciVerse ScienceDirect

journal homepage:

www.elsevier.com/locate/jbiomech

www.JBiomech.com

Journal of Biomechanics

0021-9290/$ - see front matter & 2012 Published by Elsevier Ltd. http://dx.doi.org/10.1016/j.jbiomech.2012.08.020

n

Corresponding author. Tel.: þ49 234 32 28511. E-mail address: jeanette.hussong@rub.de (J. Hussong).

(2)

Matsui et al. (1998)

, who observed that PCL and mucus are equally

fast together but that PCL transport reduces by more than 80% after

mucus removal, leading to a hypothesis that PCL is mainly driven by

frictional coupling with the mucus. Based upon this concept,

mechanistic studies on transport characteristics on surfaces of

explanted mammalian airways have carefully avoided the removal

of the mucus layer (

Henning et al., 2008

;

Ryser et al., 2007

). On the

contrary, in the absence of mucus, effective ﬂuid transport in the

amphibian upper digestive tract has been shown to generate ﬂow

up to several hundreds of micrometers above its surface when

submerged in frog Ringer buffer (

Wilson et al., 1975

). Recently, rapid

cilia-driven bead transport was measured over dissected and

mucus-freed trachea sections (see

Fig. 1

b and c) of mice (

Klein

et al., 2009

;

K ¨onig et al., 2009

).

In the present study, particle tracking velocimetry (PTV)

measurements of the bead transport as done by

K ¨onig et al.

(2009)

and

Klein et al. (2009)

are complemented with

measure-ments of the ﬂuid transport by means of microscopic particle

image velocimetry (

m

PIV) measurements. Quantitative, spatially

resolved velocity ﬁelds at micron scale can be obtained by this

measurement technique (

Adrian, 1984

;

Dudderar and Simpkins,

1977

;

Willert and Gharib, 1991

). The

m

PIV measurement technique

has recently been applied to visualize blood ﬂows in biological

systems (

Hove et al., 2003

;

Vennemann et al., 2006

;

Groenendijk

et al., 2008

;

Poelma et al., 2008

;

Nesbitt et al., 2009

;

Poelma et al.,

2010

). Our results show that not only a signiﬁcant transport of

beads (

Klein et al., 2009

;

K ¨onig et al., 2009

) but also a signiﬁcant

transport of ﬂuid takes place in the absence of mucus. Hence, we

hypothesize that directed ﬂuid transport on mammalian airway

surfaces may not require the presence of a mucus layer and,

instead, can be directly achieved in a Newtonian ﬂuid with the

characteristics of the periciliary ﬂuid.

2. Materials and methods 2.1. Animals

Wild-type mice of the lineage C57Bl/6 were terminated by inhalation of isofluorane, and the trachea was dissected immediately. The trachea, a tube of approximately 5 mm in length and 2 mm in diameter, was cut open along the entire length of the dorsal wall. The trachea immediately assumed a half-pipe shape due to residual stresses in the closed cartilage structures. The half-pipe shape was retained during the measurements. The sample was cleaned with a HEPES-buffered Ringer solution and fixed in a Petri dish with two preparation needles. The Petri dish was filled with 1.5 ml heated HEPES-Ringer and placed on a heating plate. The temperature of the buffer was kept constant at 30 1C during the whole preparation and measurement procedure. Explanted tracheae survived and maintained cellular functions for more than 4 h under such conditions.

2.2. Tracer particles

Since we performed measurements from a head-on view, background noise of the underlying epithelium was avoided by using ﬂuorescence microscopy making use of rhodamine-B ﬂuorescently labeled polyethylene-glycol (PEG) coated poly-styrene particles (Microparticles GmbH). These particles have a nominal diameter of 0:56

m

m and a specific gravity of 1.05, so they remain in suspension and follow the flow. The PEG particle coating makes the tracer surface hydrophilic. During previous experiments in the embryonic chicken hearts, no interaction between the biological system and the PEG-coated polystyrene tracers was observed (Vennemann et al., 2007;Groenendijk et al., 2008;Poelma et al., 2008,2010). In the present study, the suspended tracer particles at a volume concentration of approximately 0.003 vol% did not excite any noticeable mucus production during the measurements. In recent studies (Klein et al., 2009;K önig et al., 2009), coordinated cilia motion was observed using Protein-G-coated ferrite-polystyrene beads of specific gravity of 1.37 (Dynabeads Protein G, DYNAL Invitrogen). A Protein G coating makes the heavy beads non-toxic for biological systems. In this study the same kind of beads of 2:8

m

m nominal diameter were used (0.003 vol%). Before the measurements both the rhodamine-B-coated flow tracers and the DYNAL beads were added to the fluid in which the sample was fixed.

Fig. 1. Airway cilia (160–200) grow on one ciliated cell (Crystal et al., 1997). (a) Sketch of in vivo airway epithelium with periciliary liquid (PCL) and mucus; (b) sketch of airway epithelium after mucus removal as used in this study; (c) REM picture of a mouse trachea epithelium. Only one-third of all trachea cells in a wild-type mouse are ciliated cells (Klein et al., 2009).

Fig. 2. (a) Schematic drawing of the experimental set-up. An upright fluorescence microscope was used to perform measurements of the cilia induced bead and fluid transport. Two different particle groups are present in the fluid so that bright-field recordings of the bead transport and

m

PIV measurements of the fluid transport can be taken quasi-simultaneously. A 12-bit double-frame CCD-camera records the Protein G-coated ferrite-polystyrene beads in bright-field mode and the fluorescence signal of the rhodamine-B-coated tracers in PIV mode, respectively. The set-up is equipped with a dual-cavity laser, which illuminates the sample in the Petri dish. The laser, the camera, and the data acquisition and storage are synchronized by a PC internal trigger and timing unit. (b) Time line of the experimental protocol. The time protocol started at the moment of the mouse termination and is referred to as the zero reference time. The last measurement was completed after 91 min from the start of the protocol.

(3)

2.3. Experimental set-up

Fig. 2a shows a schematic drawing of the experimental set-up consisting of an upright fluorescence microscope (NIKON ECLIPSE FN1) and a PIV measurement system. The Petri dish with the fixed trachea sample was clamped onto a heating plate unit (Bioptechs, Butler, PA, USA), which was placed in a machined sample holder on the microscope table, so that the Petri dish, the heating plate, and the microscope table formed a fixed unit. The heating unit was equipped with a temperature sensor that was placed into the buffer solution during the experi-ments to ensure a constant liquid temperature in the Petri dish. The measured temperature regulated the heating plate in a feedback loop. A water dipping lens (Nikon CFI Apo 40XW NIR) of 40 magnification and of a numerical aperture NA¼ 0.8 was used so that the observation area, commonly called the field of view and hence on referred to as FOV, was 222 168

m

m2_{. The focal plane for this}

objective is situated at a working distance WD ¼3.5 mm with a depth of ﬁeld of 3

m

m. The measurement volume could be illuminated either with the laser light from above or with the built-in halogen light source from below for bright-ﬁeld illumination. For the epiﬂuorescence recordings, the sample was illuminated by a frequency-doubled Nd:YAG laser (Solo PIV III, New Wave), and the pictures were recorded with a 12-bit double-frame CCD-camera (Imager Intense, LaVision) at a 4.95 Hz repetition rate. The laser, the camera, the data acquisition, and the data storage were synchronized by a trigger and timing unit (PTU-8, LaVision GmbH) built in the PC.

2.4. Experimental procedure

The measurements on living cilia required a precise and repeatable timing protocol, which starts at the moment of the mouse termination and is referred to from here on as the zero reference time. Within the following 20 min the trachea was prepared, positioned in the Petri dish, and submerged in Ringer buffer solution. The Petri dish was then fixed on the heating plate under the microscope. Tracers and beads were added and mixed with the Petri dish liquid 2 min before the first measurement to ensure a homogeneous particle distribution. Forty minutes after the zero reference time, the ciliary beat has reached its baseline beat frequency of approximately 9 Hz (K önig et al., 2009). The first bright-field pictures are taken close to the epithelium surface for the PTV evaluation of the bead transport. Immediately afterwards, laser-illuminated images of the fluores-cent fluid tracers are acquired for the determination of the PCL transport. The time line is summarized inFig. 2b. Measurements were carried out with eight different mice. Three identical measurement series were performed on each trachea sample. Each measurement series consisted of recordings in eight measurement planes. In the first measurement plane the motion of the heavy beads on the epithelium surface was recorded, hence on referred to as the reference plane; the fluid motion was recorded in subsequent measurement planes, which were taken at distances of 5, 10, 15, 25, 35, 55 and 75

m

m from the reference plane. During the first measurement series the bead and fluid transport for the baseline ciliary beat (basal condition) was recorded. Five minutes before the second measurement series the beat frequency was stimulated by addition of muscarine with a final concentration of 104_{M. The final measurement series was performed 3 min after}

ATP was added to the buffer solution at a concentration of 104

M. ATP stimulation of the ciliary beat served to test the full functionality and response of the cilia at the end of the measurement procedure. In this way measurements on defect samples or samples that experienced failure or damage during the measurements could be detected and excluded from the results a priori.

2.5. Image processing and data analysis

PCL transport rates as plotted inFig. 4are based on the median velocities extracted from the whole FOV of each measurement plane. To determine velocity proﬁles above ciliated regions, ciliated cell positions relative to the focal plane were determined by identifying in-focus tracers attached to cilia. Subsequently, the relative distances between the selected ciliated regions and the measurement planes were reconstructed for ciliated regions of 5

m

m height difference as indicated on the left hand side ofFig. 5. The resulting velocity proﬁles above ciliated regions are shown on the right hand side ofFig. 5.

Conventional wall positioning methods use an extrapolation of the a priori known velocity profile to the zero velocity position (Stone et al., 2002; Vennemann et al., 2007;Poelma et al., 2008;Rossi et al., 2009,2010). This method is not applicable in our case, since the epithelium surface is covered with beating cilia, resulting in an unknown velocity profile close to the surface. We verified that for all measurements out-of-plane height differences of the epithelium relative to the measurement plane were below 15

m

m over the whole FOV of 222 168

m

m2_.

Using the Matlab image processing toolkit, a simple two-step PTV code was used to reconstruct the instantaneous bead displacements. Two hundred bright-ﬁeld recordings were taken, and the bead transport was determined from the median value of all measured bead displacements. The

m

PIV image evaluation procedure was done with a commercial code (DAVIS 7.2, LaVision GmbH). It included image preprocessing steps to enhance the visibility of the particle images

with respect to the background image; the determination of the ﬂuid velocities via PIV evaluation and several post-processing steps, to eliminate erroneous velocity vectors (see Hussong, 2011). A vector spacing of approximately 5

m

m was achieved, which corresponds to a velocity information density of four to nine vectors over one epithelium cell.

3. Results

3.1. Spatial distribution of bead and PCL velocities

Fig. 3

a shows a head-on view of a representative epithelial

area of a mouse trachea. In this preprocessed bright-ﬁeld image,

bright regions contain beating cilia, while dark regions appear to

be free of cilia. Since a single cell typically extends over 812

m

m,

one ciliated cell group outlined in white is estimated to contain

0

50

100

150

200

0

50

100

150

Fig. 3. (a) A preprocessed bright-field picture of a typical ciliated epithelial surface in a mouse trachea is shown. Ciliated cell groups are outlined in white; cilia free regions appear black. The regions labeled ‘1’ to ‘3’ are discussed inSection 3.1. (b) Discrete locations of bead positions with a specific gravity of 1.3 are indicated by red dots, and their corresponding pathlines are indicated in blue. The fluid velocity field at a plane 15

m

m above the epithelium surface is shown in black. (c) Vector ﬁeld and the contour plot of the in-plane ﬂuid velocities 15

m

m above the epithelium surface. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of this article.)

(4)

approximately 2–12 cells. A typical PTV result of beads on the

surface of the epithelium is shown in

Fig. 3

b. Bead positions, as

they were located in successive frames, are shown by red dots,

and their corresponding pathlines are drawn in blue. In this PTV

result, beads were observed to partially follow speciﬁc pathlines.

We anticipate that this is due to bead pathlines that tend to

merge as shown in

Fig. 3

b. The time-averaged ﬂuid velocity ﬁeld

15 m

m above the epithelium surface is added in

Fig. 3

b to

qualitatively compare bead and ﬂuid transport. The net transport

of beads and ﬂuid is directionally consistent with only some local

deviations. The time-averaged ﬂuid velocity ﬁeld that results from

the PIV analysis of a measurement taken 15

m

m above the

identical trachea section is shown in

Fig. 3

c. The velocity

magni-tudes are presented by colour-coded contour levels. The outlines

of ciliated cell groups as shown in

Fig. 3

a have been superimposed

on

Fig. 3

c to enable visual comparison of regions of higher

velocities and the presence of beating cilia. The average velocity

of the vector ﬁeld shown in

Fig. 3

c is 46

m

m=s. Three distinctive

regions are marked in both

Fig. 3

a and c with numbers one to

three. Maximum velocities of 141

m

m=s were measured at

posi-tion one, where a larger region of approximately 12 actively

beating ciliated cells was identiﬁed. Below-average velocities of

approximately 20

m

m=s were measured at region two, where no

beating cilia are located. An intermediate region of scattered,

ciliated cells was identiﬁed at position three, where a slight

velocity increase to 5070

m

m=s was observed compared to the

mean velocity.

3.2. Bead and PCL transport

The box plot shown in

Fig. 4

summarizes the main results of PCL

and bead transport rates by the ciliated epithelium of mice tracheae.

Signiﬁcance tests were performed with SPSS based on eight trachea

samples (n ¼ 8). The statistical analysis included a global Friedman

test followed by a Wilcoxon test to compare selected points in time.

For each trachea sample, one median bead velocity and two median

ﬂuid velocities are included, where the median is taken across the

entire FOV at three experimental conditions: (i) before stimulation

of the cilia beat (basal), (ii) after addition of the stimulant muscarine

(M), and (iii) after addition of the stimulant ATP (ATP) as labeled

below the plot. The stimulation process is described in the Methods

section. Measured bead velocities are shown on the left side in

Fig. 4

,

represented by blue-colored box plots. In the middle and on the

right side the ﬂuid velocity measurements are displayed in orange

and green in planes at approximately 2–3 and 9–13 cilia lengths

above the epithelium, respectively. Stimulated bead and ﬂuid

transport rates were signiﬁcantly higher than the corresponding

basal transport (po0:05). The addition of both stimuli, namely

muscarine and ATP, resulted in a transport rate not signiﬁcantly

distinguishable from each other (p 4 0:05). Median bead velocities

of approximately 25

m

m=s for the basal case and 110

m

m=s for the

stimulated cases were measured at the surface of the epithelium.

Median ﬂuid velocities of approximately 50

m

m=s and 150

m

m=s

were measured before and after stimulation at 2–3 cilia lengths

above the epithelium. Thus, the stimuli produced a bead velocity

increase by a factor of approximately four and a ﬂuid transport

increase by a factor of approximately three as compared with the

basal condition. The effect of the cilia beat on the ﬂuid further away

from the epithelium was investigated by performing a signiﬁcance

test between the ﬂuid transport measured approximately at 2–3 and

9–13 cilia lengths above the surface. A signiﬁcant decrease of 14–

33% in the stimulated ﬂuid transport is evident to transport rates of

105120

m

m=s medial in the highest plane. The difference in ﬂuid

transport at these two planes was not statistically signiﬁcant in the

unstimulated (basal) case due to variability in the data between

samples. (The apparent statistical outlier in measured basal velocity

9–13 cilia lengths above the epithelium shown in

Fig. 4

is included

in the calculation.)

The Reynolds number of the bead transport is in the order of

Re

b

¼

v

b

D

b

=

n

Oð10

5

10

4

Þ, where D

b

refers to the nominal bead

diameter,

n

is the kinematic viscosity of the ﬂuid and v

b

are the

median bead velocities of the basal and stimulated ﬂow. The relative

importance of gravity and lift on the bead motion compared to

viscous drag can be estimated with the help of the measured

time-averaged bead and PCL transport rates. Lift forces occur in the

presence of velocity-gradient-induced bead rotation (

Happel and

Brenner, 1973

). In a stationary shear ﬂow the ratio between lift

and viscous drag forces can be estimated to (

Saffman, 1965

;

Maxey

and Riley, 1983

): F

lift

=F

d

Oð

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

ðD

2b

=

n

Þ

du=dy

q

Oð10

3

Þ, where the

local velocity gradient du=dy is estimated from the PCL velocity

difference between uðz ¼ 0Þ 0 at the non-ciliated epithelium

surface and uðz ¼ 15

m

mÞ 50150

m

m=s as acquired from

m

PIV

measurements.

3.3. PCL velocities above single ciliated cells

The right hand side of

Fig. 5

reveals further details

regard-ing the epithelium-parallel velocity as a function of the distance

from the surface. Two velocity proﬁles are shown as they were

measured over ﬁve single ciliated cells of one representative

mouse before (black curve) and after the cilia beat was stimulated

with muscarine (red curve). The uncertainty of the

epithelium-normal location of the measurement plane is represented by the

vertical error bars. The horizontal error bars show the range of

ﬂuid velocities measured in each plane above the identiﬁed

ciliated cells. No ﬂuid velocities could be measured in planes of

distances less than 10

m

m to the ciliated cells since the signal of

Fig. 4. The box plots show the median (bold line), the 25% and 75% percentiles (box dimensions) and the extreme deviations (vertical lines) of the measurement results. An outlier value is marked by a circle. Shown are the measurement results of bead and ﬂuid transport. The measured bead velocities are represented in blue, and the ﬂuid velocities at planes 15

m

m and 75

m

m above the epithelium are depicted by orange and green bars, respectively. For each group the basal velocities and the velocities after successive muscarine and ATP stimulation are illustrated from left to right by a separate bar. The stimulation of the ciliary beat with muscarine and ATP leads to a significant increase in bead and fluid velocities with respect to the corresponding basal velocities (npr0:05 vs. basal value). The decrease in fluid transport between the measurement planes 15

m

m and 75

m

m above the epithelium surface is signiﬁcant for the stimulated transport (#pr0:05 vs. plane 15

m

m above epithelium). Fluid transport rates measured by PIV in a plane 15

m

m above epithelium were signiﬁcantly higher than the bead transport rates (þ pr0:05 vs. corresponding PTV) before and after stimulation. Statistical analysis: global Friedman test followed by a Wilcoxon test to compare selected points in time. (For interpretation of the references to color in this ﬁgure legend, the reader is referred to the web version of this article.)

(5)

tracers attached to the cilia was stronger than the signal of

suspended particles, thereby spreading the correlation peak and

resulting in erroneous vectors. Supporting the composite

statis-tics in

Fig. 4

, a noticeable decrease in ﬂuid transport velocity was

observed in the measurement plane at 75

m

m from the

epithe-lium surface as compared to the ﬂuid velocity 15

m

m above the

epithelium. The velocity proﬁles in

Fig. 5

show that net ﬂuid

transport for this sample continually decreases with increasing

distance to the epithelium surface.

4. Discussion and conclusions

In the present study, time averaged cilia-induced ﬂuid and

bead transport rates were measured quasi-simultaneously above

mucus-freed, ciliated-surface areas of the mouse trachea. The

cilia-driven ﬂuid transport was measured directly by means of

m

PIV measurements with neutrally buoyant tracers.

While heavy beads provide only insight in the wall-close

transport process, direct measurement techniques such as the

PIV technique are necessary to correctly quantify the spatial

distribution of the PCL transport. In the present study, PCL

transport rates were shown to decrease with increasing spacing

to the tracheal epithelium.

Median bead velocities of approximately 25

m

m=s basal and

110 m

m=s stimulated were measured at the surface of the

epithelium. Time averaged ﬂuid velocities of approximately

50 m

m=s and 150

m

m=s stimulated were measured in a plane

parallel to and above the epithelium surface at a distance of

approximately 23 times a cilia length. According to recent

studies, PCL transport would be initiated by the frictional

inter-action with the mucus layer (

Knowles and Boucher, 2002

), and

therefore no signiﬁcant PCL transport would take place in planes

below the ciliary tips after mucus removal (

Matsui et al., 1998

).

However, our results show that both effective bead transport at

the surface of the epithelium and effective PCL transport 15

m

above the epithelium can take place in the absence of mucus.

Mean bead transport velocities are estimated to be two orders

of magnitude larger compared to bead sedimentation velocities.

Thus, in non-ciliated regions beads are expected to closely follow

the PCL motion. We therefore conjecture that in non-ciliated

epithelium regions the presence of bead transport also indicates

the presence of PCL transport.

Based on our results, we have shown that PCL transport can be

directly driven by the cilia beat and that the PCL motion may be

capable of driving bead transport by ﬂuid drag.

Conﬂict of interest statement

The authors wish to conﬁrm that there are no known conﬂicts

of interest associated with this publication and there has been no

signiﬁcant ﬁnancial support for this work that could have

inﬂu-enced its outcome.

Acknowledgments

This research is sponsored by the Sixth EU framework program

ARTIC (STRP 033274) and the Excellence Cluster Cardio-Pulmonary

System.

References

Adrian, R.J., 1984. Scattering particle characteristics and their effect on pulsed laser measurements of ﬂuid ﬂow: speckle velocimetry vs particle image velocimetry. Applied Optics 23 (11), 1690–1691.

Braiman, A., Priel, Z., 2008. Efﬁcient mucociliary transport relies on efﬁcient regulation of ciliary beating. Respiratory Physiology & Neurobiology 163 (1–3), 202–207.

Crystal, R.G., West, J.B., Weiber, E.R., Barens, P.J., 1997. The Lung: Scientiﬁc Foundations, 2nd ed. Lippincott-Raven, Philadelphia.

Donaldson, S.H., Bennett, W.D., Zeman, K.L., Knowles, M.R., Tarran, R., Boucher, R.C., 2006. Mucus clearance and lung function in cystic ﬁbrosis with hyper-tonic saline. New England Journal of Medicine 354 (3), 241–250.

Donaldson, S.H., Boucher, R.C., 2007. Sodium channels and cystic ﬁbrosis. Chest 132 (5), 1631–1636.

Dudderar, T.D., Simpkins, P.G., 1977. Laser speckle photography in a ﬂuid medium. Nature 270 (5632), 45–47.

Fahy, J.V., Dickey, B.F., 2010. Airway mucus function and dysfunction. New England Journal of Medicine 363 (23), 2233–2247.

Francis, R.J.B., Chatterjee, B., Loges, N.T., Zentgraf, H., Omran, H., Lo, C.W., 2009. Initiation and maturation of cilia-generated ﬂow in newborn and postnatal mouse airway. American Journal of Physiology Lung Cellular and Molecular Physiology 296 (6), L1067–L1075.

Groenendijk, B.C.W., Stekelenburg-de, V.S., Vennemann, P., Wladimiroff, J.W., Nieuwstadt, F.T.M., Lindken, R., Westerweel, J., Hierck, B.P., Ursem, N.T.C., Poelmann, R.E., 2008. The endothelin-1 pathway and the development of cardiovascular defects in the haemodynamically challenged chicken embryo. Journal of Vascular Research 45, 54–68.

Happel, J., Brenner, H., 1973. Low Reynolds Number Hydrodynamics. Noordhoff International Publishing, Leyden.

Henning, A., Schneider, M., Bur, M., Blank, F., Gehr, P., Lehr, C.-M., 2008. Embryonic chicken trachea as a new in vitro model for the investigation of mucociliary particle clearance in the airways. AAPS PharmSciTech 9 (2), 521–527. Hove, J.R., Koster, R.W., Forouhar, A.S., Acevedo-Bolton, G., Fraser, S.E., Gharib, M.,

2003. Intracardiac ﬂuid forces are an essential epigenetic factor for embryonic cardiogenesis. Nature 421, 172–177.

Hussong, J., 2011. Flow Induced by Beating Artiﬁcial Cilia: An Experimental and Numerical Study. Ph.D. Thesis, Delft University of Technology.

Klein, M.K., Haberberger, R.V., Hartmann, P., Faulhammer, P., Lips, K.S., Krain, B., Wess, J., Kummer, W., K ¨onig, P., 2009. Muscarinic receptor subtypes in cilia-driven transport

0

20

40

60

80

100

120

140

160

180

0

20

40

60

80 Velocity [µm/s]

Distance to ciliated cell [

µm]

Without stimulation

After stimulation

Fig. 5. Left: reconstruction of the distances between measurement planes and ciliated cells. Right: two epithelium-normal profiles of the in-plane velocity component were measured over five ciliated cells of one trachea. The net fluid transport was measured before (black line), and after stimulation of the ciliary beat with muscarine (red line). The results show a continuous decrease in net fluid transport with increasing distance to the epithelium. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

(6)

and airway epithelial development. European Respiratory Journal 33 (5), 1113–1121.

Knowles, M.R., Boucher, R.C., 2002. Mucus clearance as a primary innate defense mechanism for mammalian airways. Journal of Clinical Investigation 109 (5), 571–577.

K ¨onig, P., Krain, B., Krasteva, G., Kummer, W., 2009. Serotonin increases cilia-driven particle transport via an acetylcholine-independent pathway in the mouse trachea. PloS ONE 4 (3), e4938.

Mall, M.A., 2008. Role of cilia, mucus, and airway surface liquid in mucociliary dysfunction: lessons from mouse models. Journal of Aerosol Medicine And Pulmonary Drug Delivery 21 (1), 13–24.

Matsui, H., Randell, S.H., Peretti, S.W., Davis, C.W., Boucher, C., 1998. Coordinated clearance of periciliary liquid and mucus from airway surfaces. Journal of Clinical Investigation 102 (6), 1125–1131.

Maxey, M.R., Riley, J.J., 1983. Equation of motion for a small rigid sphere in a nonuniform ﬂow. Physics of Fluids 26 (4), 883–889.

M ¨oller, W., H ¨außinger, K., Ziegler-Heitbrock, L., Heyder, J., 2006. Mucociliary and long-term particle clearance in airways of patients with immotile cilia. Respiratory Research 7 (1), 10.

Mossberg, B., Afzelius, B.A., Eliasson, R., Camner, P., 1978. On the pathogenesis of obstructive lung disease. A study on the immotile-cilia syndrome. Scandina-vian Journal of Respiratory Diseases 59 (2), 55–65.

Nesbitt, W.S., Westein, E., Tovar-Lopez, F.J., Tolouei, E., Mitchell, A., Fu, J., Carberry, J., Fouras, A., Jackson, S.P., 2009. A shear gradient-dependent platelet aggrega-tion mechanism drives thrombus formaaggrega-tion. Nature Medicine 15 (6), 665–673. Poelma, C., Van der Heiden, K., Hierck, B.P., Poelmann, R.E., Westerweel, J., 2010. Measurements of the wall shear stress distribution in the outﬂow tract of an embryonic chicken heart. Journal of the Royal Society Interface 6,7 (42), 91–103.

Poelma, C., Vennemann, P., Lindken, R., Westerweel, J., 2008. In vivo blood ﬂow and wall shear stress measurements in the vitelline network. Experiments in Fluids 45 (4), 703–713.

Rossi, M., Lindken, R., Hierck, B.P., Westerweel, J., 2009. Tapered microﬂuidic chip for the study of biochemical and mechanical response at subcellular level of endothelial cells to shear ﬂow. Lab Chip 9 (10), 1403–1411.

Rossi, M., Lindken, R., Westerweel, J., 2010. Optimization of multiplane

m

PIV for wall shear stress and wall topography characterization. Experiments in Fluids 48 (2), 211–223.

Ryser, M., Burn, A., Wessel, T., Frenz, M., Rika, J., 2007. Functional imaging of mucociliary phenomena. European Biophysics Journal 37 (1), 35–54. Saffman, P.G., 1965. The lift on a small sphere in a slow shear ﬂow. Journal of Fluid

Mechanics 22, 385–400.

Sleigh, M.A., Blake, J.R., Liron, N., 1988. The propulsion of mucus by cilia. American Review of Respiratory Disease 137 (3), 726–741.

Stone, S.W., Meinhart, C.D., Wereley, S.T., 2002. A microﬂuidic-based nanoscope. Experiments in Fluids 33 (5), 613–619.

Vennemann, P., Kiger, K.T., Lindken, R., Groenendijk, B.C.W., Stekelenburg-de Vos, S., ten Hagen, T.L.M., Ursem, N.T.C., Poelmann, R.E., Westerweel, J., Hierck, B.P., 2006. In vivo micro particle image velocimetry measurements of blood-plasma in the embryonic avian heart. Journal of Biomechanics 39 (7), 1191–1200.

Vennemann, P., Lindken, R., Westerweel, J., 2007. In vivo whole-ﬁeld blood velocity measurement techniques. Experiments in Fluids 42 (4), 495–511. Willert, C.E., Gharib, M., 1991. Digital particle image velocimetry. Experiments in

Fluids 10 (4), 181–193.

Wilson, G.B., Jahn, T., Fonseca, J., 1975. Studies on ciliary beating of frog pharyngeal epithelium in vitro i. Isolation and ciliary beat of single cells. Transactions of the American Microscopical Society 94, 43–57.