the neutrophil lungs: haemodynamics, cell andtion (fig 2). the regional retention of radiolabelled...

Thorax1993;48:79-88 79~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

British Association for Lung Research review series Series editor: Duncan F Rogers

New perspectives on basic mechanisms in lung disease* 2Neutrophil traffic in the lungs: role ofNeutrophil traffic in the lungs: role ofhaemodynamics, cell adhesion, and deformability

W MacNee, C Selby

There has been increasing interest in recentyears in the traffic of neutrophils in thelungs,' prompted by several factors.Neutrophils are the first cells to arrive at sitesof infection or inflammation, and are presentin the lungs in increased numbers in manyconditions of acute and chronic lung injury.45They have thus been implicated in the patho-genesis of many lung diseases through therelease, from activated cells, of reactive oxy-gen intermediates6 and proteolytic enzymes.'Uniquely, the pulmonary capillary bedreceives all of the circulating neutrophils andalso contains a pool of temporarily non-circu-lating cells.8 Influenced by various stimuli,these cells can either rejoin the circulatingpool or migrate to an inflammatory site. It isin the pulmonary capillaries that neutrophilscan interact with endothelial cells, with whichthey are in close contact. In addition, closeproximity to the alveoli allows the transfer ofmessages between the air and vascular spaces.

In this review we will assess the influence ofthe factors that govern neutrophil traffic.

Figure 1 Scanning electron micrograph of a latex cast of the alveolar capillary bed of arat lung, showing the interior view of the alveolar meshwork. The basic component isclearly cylindrical. The human pulmonary capillary bed is considered to have a similararrangement,'0 but has yet to be so well illustrated. The bar represents IO pm. Reproducedfrom Guntheroth et al 13 by courtesy of the American Physiological Society.

Such knowledge is necessary for the completeunderstanding of conditions of lung inflam-mation and injury.

Anatomical and morphometric factorsIn 1661 Malpighi9 described the pulmonarymicrocirculation as a network of tubular cap-illaries (derived from the Latin capillaris: ofhair, hair like). In his book on the morphome-try of the human lung Weibel'° extended thisobservation by describing a hexagonalarrangement of interconnecting tubules. Thisview was disputed by Fung and Sobin," 1"whose description of the microcirculation,based largely on light microscopy and thetheory of fluid mechanics, was of sheets ofendothelium separated by posts, betweenwhich blood flowed, like "a parking garagewith floor and ceiling, and intervening posts."More recent studies, using latex casts andscanning electron microscopy, show convinc-ingly that the alveolar microcirculation iscomposed of tightly matted intersectingtubules, rather than "sheets" (fig 1).'3 Thisconfiguration complicates the path taken bycirculating neutrophils from arteriole tovenule. The surface area of the alveolar capil-laries in the human lung has been estimatedto be of the order of 90 in2, consisting of 296x 106 alveoli and 277 x 109 capillary seg-ments-that is, 1000 capillary segments peralveolus.'4 Hogg' calculated that the averagepathway for a neutrophil from arteriole tovenule has 60 capillary segments.

There is a discrepancy between the diame-ter of the circulating neutrophils, mean 7 03(range 5-8) pm,'5 and that of the pulmonarycapillary segments, mean 5 (range 1-16)um) .'4 Similar dimensions have recently beenconfirmed by morphometry of neutrophilsand capillaries in situ in resected humanlungs.'6 Capillary diameters are, however,influenced by alveolar pressure, which givesrise to zones of differing capillary recruit-ment.'7 Accordingly, the mean capillary dia-meter has been calculated to be 1-99 pm inzone II conditions and 5-78 pm in zone III inthe rat lung.'8

Obstruction of blood flow, by neutrophilsthat plug capillaries and impede the passageof erythrocytes, is well established in the sys-temic microcirculation and is enhanced by areduction in blood flow.'9 This may be impor-tant in the pathogenesis of many ischaemic

Unit of RespiratoryMedicine,Department ofMedicine, CityHospital, EdinburghEH10 5SBW MacNeeC SelbyReprint requests to:Dr W MacNee

79Thorax 1993;48:79-88

on June 7, 2020 by guest. Protected by copyright.

http://thorax.bmj.com

/T

horax: first published as 10.1136/thx.48.1.79 on 1 January 1993. Dow

nloaded from

http://thorax.bmj.com/

MacNee, Selby

diseases.20 The transit of neutrophils throughthe pulmonary microcirculation is less wellcharacterised, and for several reasons maydiffer from that of the systemic circulation.Pressures within the pulmonary circulationare much lower than in the systemic circula-tion2' and flow is pulsatile.22 In addition, theaverage diameter of the pulmonarycapillaries'5 is probably less than that of thesystemic capillaries (6 pm)2' and the proper-ties of endothelial cells may differ betweenthe circulations. This means that the resultsof studies of neutrophil transit may also differbetween the two vascular beds.

Margination, retention, sequestration,and migrationNeutrophils are present in three separatepools within the body-a bone marrow pool,an intravascular pool, and a pool within thetissues. Maturation within the bone marrowand release of neutrophils from it are not thesubject of this review, but clearly factors thatalter the release of neutrophils from the bonemarrow will affect the intravascular pool ofneutrophils.2425 The fate of neutrophils in thetissues is not well understood, but probablyincludes the process of apoptosis or "pro-grammed cell death." 26

Neutrophils spend a relatively short periodof their total life span within the intravascularspace (T/2 = 7 hours). Early studies of neu-trophils radiolabelled with diisopropylfluoro-phosphate (DF'2P, reviewed in refs 1 and 2)indicated that more than half of these cellsrapidly disappeared from the blood on rein-jection, forming a non-circulating pool ofcells, which did not relate to the removal ofcells damaged as a result of the labelling pro-cedure.25 Subsequent studies suggested thatthe circulating and the non-circulating poolscontained roughly equal numbers of cells.8This non-circulating pool of neutrophils, withthe pool of mature and immature cells withinthe marrow, can be released in times ofstress.27

Cohnheim28 first noticed that neutrophilsappeared to "marginate" to vessel walls, whileerythrocytes remained in the central stream ofblood. This led to the idea that the intra-vascular non-circulating neutrophil pool con-sisted of marginating cells, temporarilywithdrawn from the circulation. Schmid-Schonbein and coworkers'9 have shown that94% of circulating neutrophils marginate inthe postcapillary vessels in the systemic circu-lation, which occurs at the step change in ves-sel diameter between the capillary and thepostcapillary venule, and has been shown tobe flow dependent.29 Whether true neutrophilmargination occurs in the pulmonary circula-tion remains a subject of debate.' 2More recent studies (reviewed in refs 1 and

2), using both labelled and unlabelled cells,supported the idea that the lungs contained alarge pool of non-circulating neutrophils thatwere retained or sequestered within the pul-monary circulation. There is certainly evi-dence, however, that some harvesting30 and

labelling procedures3' may be detrimental tocell function. Such ex vivo activation or evencell injury may account for the variation inthe patterns of neutrophil kinetics observedafter the reinjection of radiolabelled neu-trophils in some studies in patients. Indeed,some authors claim that any retention ofradiolabelled neutrophils in the pulmonaryvasculature is due entirely to ex vivo cellactivation or injury.32 The same authors havecalculated that the so called "marginatingneutrophil pool" accounted for 54% of thetotal blood granulocyte pool and that the lungscontained 33% of the total marginating pool.33The debate over whether the lungs contain asubstantial pool of non-circulating neutrophilsis compounded by the misuse of the termmargination; this correctly refers to neutro-phils, located predominantly in the post-capillary venules in the systemic circulation,that roll slowly along vessel walls.2934 Theterm should not be used synonymously withneutrophil retention, which is the number ofcells retained as a percentage of the cellsdelivered to an organ. We prefer to use theterm sequestration to describe cells other thanthe freely circulating pool of neutrophilswithin the pulmonary vasculature. Theseinclude cells that are truly marginating, thosethat are trapped, those adherent to the endo-thelium, and those slowly moving through themicrovasculature. The process of intravascu-lar neutrophil sequestration in the lungs iscritical to the subsequent migration of thesecells, because to migrate cells in transit in thecirculation must stop if they are to adhere tothe endothelium before diapedesis.Among the most convincing evidence that

the lungs influence the passage of circulatingneutrophils is that obtained by studies ofunlabelled cells, where interventions such asthe Valsalva manoeuvre cause increasedsequestration of neutrophils in the pulmonaryvasculature.35 These findings are supportedby morphometric data on animals, wheremore neutrophils have been foundsequestered in the lungs than in the circulat-ing pool.36 Additional support comes fromdirect intravital observation in animal lungs ofthe delay of a proportion of neutrophils intransit within the pulmonary microvascula-ture.37 Moreover, the study of the transit ofradiolabelled neutrophils in the lungs seemsjustified because after their reinjection radio-labelled cells are distributed in the pulmonaryvasculature in the same way as unlabelledcells,36 and respond to signals to rejoin thecirculating pool in a fashion similar to that ofnative unlabelled cells.'8-42

Haemodynamic factorsA series of experiments, conducted over thepast 10 years largely in animals and principal-ly by Hogg's group in Vancouver,' haveexamined the influence of local haemody-namics on the sequestration of radiolabelledneutrophils in the lungs by comparing theirtransit with that of radiolabelled erythrocytes.Initial experiments in the dog suggested that

80



/T


nloaded from


Neutrophil traffic in the lungs: role of haemodynamics, cell adhesion, and deformability

Right ventricle

o 1 1 1 In neutrophils+ 99mTc erythrocytes

20 40

Time (seconds)B

Left lung

0 1'11n neutrophils+ 99mTc erythrocytes

20 40 60

Time (seconds)Figure 2 First pass time-activity curves from regions of interest assigned to the ightventricle (A) and left lung (B) after a bolus injection of indium-l11 labelled neutrophilsand technetium-99m labelled erythrocytes. Allfour curves are normalised to their peakcounts and represent the first few seconds after reinjection. The two cell types pass throughthe right ventricle synchronously. Neutrophils, however, are slower in their transit throughthe lung than erythrocytes.

80-90% of reinjected radiolabelled neu-

trophils were removed in a single pass

through the lungs.3839 A high initial rate ofneutrophil sequestration in the lungs has beenconfirmed in other species, such as the rab-bit36 and pig.43 In man there is also good evi-dence, obtained with a gamma camera thattracked the passage of an intravenous bolusinjection of indium-i 11 labelled neutrophilsand technetium-99m labelled erythrocytes,that neutrophils are delayed more than ery-

throcytes during their first transit through thepulmonary circulation before their recircula-tion (fig 2).The regional retention of radiolabelled

neutrophils, 10 minutes after reinjection,appears to depend on local haemodynamics,both in animals4' and in man.4445 Reducingthe cardiac output by inflating a balloon inthe inferior vena cava in the dog produced an

arteriovenous difference of unlabelled cells,fewer being present in systemic arterial blood

than in venous blood.38 These sequesteredneutrophils are subsequently released if flowis increased.39 If a reduced flow is maintainedany arteriovenous difference in unlabelledneutrophils diminishes after 10 minutes as anew equilibrium is reached and cells areadded from other pools in the body.39 Thesestudies support the concept of a dynamicequilibrium between circulating and non-circulating neutrophil pools within thepulmonary circulation. In support of thisconcept, Perlo and colleagues46 showed thatthe ratio of white blood cells to red bloodcells was greater in zone II than in zone III inisolated perfused lungs, though the increasein white blood cells was largely due toincreased lymphocytes. They concluded thatcompression of the microvasculature in zoneII may cause the alveolar vessels to act as asieve for neutrophils. Recent studies in patientsundergoing cardiac catheterisation seem toconfirm this view because increasing alveolarpressure, causing compression of the alveolarvessels, produced a transient fall in neutrophilcount in the left ventricular blood but not inthe pulmonary arterial blood, as a resultpresumably of neutrophil sequestration inthe lungs.35 Moreover, the fall in neutrophilscorrelated with the proportion of the lungthat changed from zone III to zone II.3The passage of neutrophils through the

pulmonary microcirculation depends on thebalance between the forces that tend to retardtheir progress and the dispersive forces thatact to move them through the vessels. Theretardive forces are those of adhesion andfriction between the neutrophil and theendothelial cell and those related to thedeformation of the neutrophil. Deformationof the neutrophil is necessary as it has toachieve a smaller effective diameter tosqueeze through the smaller diameter capil-laries. The dispersive forces are those relatedto the shear stress of the vessel wall, which inturn depends on the local blood velocity.47The effect of local blood velocity on neu-

trophil sequestration in the lungs can be stud-ied by relating retention of neutrophils to thetransit time of erythrocytes across the lungs.Staub and Schultz48 have shown that there isno regional variation in the distribution ofpathway lengths between different regions ofthe lung, at least in animal lungs. Thus themeasurement of regional transit time of ery-throcytes, either directly by intravitalmicroscopy49 or indirectly by obtaining time-activity curves for the transit of a bolus ofradiolabelled erythrocytes across the lungs,50provides a measure of local blood velocity. Indogs,5' rabbits,36 and man50 if markers thatbecome lodged in proportion to flow (radio-labelled microspheres or macroaggregatedalbumin) are injected intraoperatively with avolume marker (radiolabelled erythrocytes)counts on the volume marker can be dividedby counts on the flow marker in subsequentlyresected specimens of lung, and this enablesthe regional transit time of the erythrocytes tobe calculated. These studies show a regionalvariation in the transit time of erythrocytes,

A100 7

90 -

80-

70-

c 60-0" 50-'a)= 40 1

Eo 30z

20

10

0

100-

90g

80 -

. 70(A4-- 60-00 500)

, 40Eo 30z

20

10 -

81



/T


nloaded from


MacNee, Selby

with longer transit times, and hence slowerblood velocities, in the upper, more depen-dent regions.'65051 This has been confirmed inman with a gamma camera technique."50There is also a correlation between the num-ber of neutrophils retained 10 minutes afterinjection of radiolabelled cells and the ery-

throcyte transit time, so that the longer theerythrocyte transit time the more neutrophilsare retained.'6 41 44 45 52

These data seem to suggest that localblood velocity, and hence shear stress, are

important determinants of the retention ofradiolabelled neutrophils in vivo 10 minutesafter their reinjection. This may be relevant to

the increased sequestration of neutrophils instates of shock, where blood flow is reduced,which may be important in the pathogenesisof the adult respiratory distress syndrome(ARDS).5' Indeed, peripheral blood neutro-

penia, due presumably to increased neu-

trophil sequestration in the lungs, has beenshown to occur as a prodromal event inpatients at risk of developing this syndrome.54The physiological shear stresses calculated

to occur in the pulmonary microcirculation(100-500 dyne cm-2) may prevent the adher-ence of both normal and activated neu-

trophils to endothelium. Any decrease inflow, however, may alter this interaction.55

Furthermore, a reduction in the transittime of erythrocytes-for example, by treat-ment with adrenalin40-should decreaseretention of neutrophils in the lungs.Erythrocyte transit time was reduced, and theexchange rate between the pools of circulat-ing and non-circulating neutrophilsincreased, in the lungs of rabbits treated withadrenalin in regions with short erythrocytetransit times. There was no overall change,however, in neutrophil retention in thelungs.40 This can be explained by decreasedneutrophil retention in regions with shorttransit times, which was offset by the recruit-ment of underperfused capillary segmentsproduced by this treatment, in which transittimes were longer and hence neutrophilretention was greater.The site of neutrophil sequestration in the

lungs is relevant to the factors that controlsuch sequestration. The relation betweenblood velocity and neutrophil reten-

tion'641444552 seems to suggest that true mar-

gination, which is flow dependent,28 may

occur in the normal lung. The geometric con-

straints imposed on the larger neutrophils inthe smaller pulmonary capillaries, however,mean that such margination would be most

likely to occur in the postcapillary venules, as

in the systemic circulation.28 As interventionsthat reduce erythrocyte transit time, such as

infusion of adrenalin,40 smoking,44 and infu-sion of activated plasma,56 are associated withan increase in neutrophil sequestration, thissuggests that factors other than those relatedto local haemodynamics may influence thetransit of neutrophils in the lungs. Indeed,studies in man indicate that, although the 10

minute neutrophil sequestration in the lungscorrelates with erythrocyte transit time, blood

velocity does not influence the initial or firstpass sequestration.57

As described above, Doerschuk andcoworkers,36 using both morphometric tech-niques for unlabelled neutrophils and auto-radiography in studies of radiolabelled cells,have convincingly shown that labelled andunlabelled cells have a similar distribution inthe pulmonary circulation of the normalrabbit lung. In this study 74% of neutrophilsinjected intravenously were removed in thefirst passage through the lungs, and 70% ofthe injected cells could be located in thelungs, liver, or spleen at 10 minutes. Most ofthe cells in the lungs were located in thealveolar capillaries. Morphological studiesshow that the ratio of neutrophils toerythrocytes was almost 10 times greater inthe alveolar capillaries than in the extra-alveolar vessels. From these studies the authorscalculated that in the rabbit the size of the non-circulating pool of neutrophils in the lungswas at least twice that of the circulatingpool.36 This is a larger difference than thatreported previously for man, in whom the total"marginating" pool of neutrophils wassimilar in size to the circulating pool.8Indeed, some authors have calculated that thelungs contribute a relatively small percentageto the total "marginating" pool."

Using direct in vivo videomicroscopy toobserve the transit of fluorescently labelledneutrophils through a transparent window inthe periphery of the dog lung, Lien andcoworkers37 confirmed that neutrophils weresequestered exclusively in the pulmonary cap-illaries. None of the cells were delayed in thearterioles or in the venules and the authorsdid not observe any rolling or margination ofcells in the pulmonary microcirculation. Thedistribution transit times of neutrophils in thecapillaries was wide, ranging from under 2seconds to over 20 minutes. Lien et al 58 alsocalculated that only a small number ofobstructions (2%) were needed to stop most(71%) of the neutrophils in the capillaries.These neutrophils do not obstruct the pas-sage of erythrocytes because the excess ofcapillary segments allows streaming of ery-throcytes around them. Less than half of theneutrophils passed through with no delayfrom arteriole to venule. Most of the neu-trophils that were delayed in the pulmonarycapillaries stopped completely for varyinglengths of time. Very few cells moved slowlythroughout their transit."758 Although in vivomicroscopy of the lung is possible only for thesubpleural microcirculation, capillary diame-ter, which is the critical dimension, is similarin the subpleural capillaries and in the capil-lary bed in the deep lung.-'

In a further study using the same tech-nique, Lien and colleagues58 observed theeffects of changes in pressure and flow on thetransit of neutrophils in the pulmonary capil-laries of the dog lung. The transit times offluorescently labelled neutrophils in the pul-monary capillaries fell when adrenalin or

hypoxia was administered: the two proce-dures increased pulmonary arterial pressure

82



/T


nloaded from



by the same amount, adrenaline in additioncausing an increase in cardiac output.)8Balloon inflation in the vena cava, by con-trast, reduced cardiac output by 41% withoutchanging pulmonary arterial pressure andproduced an increase in the transit times ofneutrophils through the capillaries by shiftingthe distribution of transit times between thefastest and the slowest groups of cells.Muir and coworkers42 have shown an

increase in the numbers of circulating radiola-belled and unlabelled neutrophils in responseto adrenalin infusion or exercise in normalsubjects, which appeared to result from therelease of cells from a pool in the lungs.Recently, Peters et al 9 challenged the ideathat the lung contributed cells to the leucocy-tosis of exercise. Different techniques wereused in the two studies,42 59 however, to calcu-late neutrophil sequestration in the lungs, andthe exercise protocols were also different.Moreover, in the study by Muir et al 42 mea-surements were made within 15 minutes ofsimultaneous reinjection of simultaneouslyradiolabelled neutrophils and erythrocytes,whereas in the study by Peters et aP9 the twocell types were injected at different times andthe measurements were made at a later timeafter reinjection, at a time of low activitywithin the lungs.

As described above, several studies havenow shown that certain interventions thatreduce erythrocyte transit time and shouldtherefore decrease neutrophil sequestration inthe lungs paradoxically increase sequestra-tion. These studies suggest that factors otherthan local blood velocity may also influencethe transit of neutrophils in the pulmonarymicrocirculation.

Neutrophil deformabilityThe influence of the mechanical properties ofneutrophils on blood flow in the systemicmicrocirculation has been increasingly recog-nised.47606' Although present in smaller num-bers in the blood than erythrocytes,neutrophils are not easily deformed,62 whichleads to entrapment in the capillaries. If sub-sequently activated,63 with the release of oxy-gen radicals and proteases, these sequesteredcells could contribute to the pathogenesis ofischaemic diseases20 and to lung injury.5364 Asboth erythrocytes and neutrophils are largerthan the pulmonary capillaries'4 '5 the rheo-logical properties of blood cells, particularlythe ability of cells to deform in order tosqueeze through the smaller capillaries, arelikely to be important determinants of thetransit of these cells in the lung.The deformability of a cell depends on: (1)

the viscoelastic properties of the cell mem-brane; (2) the viscoelastic properties of thecell cytoplasm; and (3) the surface area-vol-ume relationship.65 Neutrophils are 700 timesless deformable than erythrocytes, though ofa similar size.65 The spherical shape of theneutrophil, however, is much less deformablethan the biconcave disc shape of the erythro-cyte. The neutrophil also contains a rigid

nucleus and its granular cytoplasm is 1000times more viscous than the cytoplasm of theerythrocyte.65 Although the two cells are simi-lar in size the neutrophil, being spherical, hastwice the volume of the erythrocyte. Bothneutrophils and erythrocytes have excess sur-face areas, greater than is necessary forenclosing their volume as a sphere. In thecase of the neutrophil there is an 84% excesssurface area,65 due to cell membrane ruffling,which allows the cell to change shape andactively squeeze through tight junctionsbetween endothelial cells as small as 1 pm indiameter. Passive deformation, which occursduring micropipette aspiration in vitro or dur-ing transit through capillaries of smallerdiameter in vivo, requires a threshold pres-sure to unruffle the membrane followed by aslower deformation of the viscous cyto-plasm.66 67The deformability of white blood cells can

be studied by several techniques. Themechanical properties of populations ofneutrophils can be studied by cell filtrationthrough a micropore membrane68-70 and theinfluence of pore size, cell deformability, andhydrodynamic factors determined.7' Theaverage dimensions of the pulmonary capil-lary segments can be mimicked in vitro bythis technique by using a micropore mem-brane with a pore diameter of 5 um and alength of 11 pum. Several filtration techniqueshave been used. Cells can be filtered at a con-stant flow and the pressure developed overtime within the filtration chamber mea-sured.68 Alternatively, the cells can be filteredat a constant pressure and a flow-time curvegenerated.72 Neutrophils can also be radiola-belled and their retention in filters assessed.In this way the effects of varying both pres-sure and flow on neutrophil retention can bestudied.7' The technique used depends on thequestions asked concerning the biophysicalproperties of the neutrophil.More direct methods have also been used

to study the deformability of individual cells.Cell aspiration with micropipettes of 5,uminternal diameter can be used to measure celldeformation time as a measure of deformabil-ity.74 Probably the best method to assess cellstiffness or deformability, however, is the"cell poker" technique. This device indentsthe cell surface in a time dependent mannerand measures both the force and the degreeof cell indentation.7576With these techniques populations of leu-

cocytes appear to vary in their stiffness,monocytes being less deformable than neu-trophils.77 The most important hydrodynamicfactor influencing the retention of neutrophilsduring filtration through a micromembrane invitro is the perfusion pressure and hence theshear stress, so that as the shear stressincreases the retention of neutrophils in a fil-ter decreases in a logarithmic fashion.7' Theshear stresses used in these studies of modelcapillaries (80-800 dyne cm-2)73 are of thesame order of magnitude as those calculatedto occur in the pulmonary microcirculation invivo (100-500 dyne cm-2),71 but are three

83



/T


nloaded from


MacNee, Selby

orders of magnitude greater than is necessaryfor inhibiting adherence of neutrophils toendothelium.5573 Five pm appears to be thecritical pore size at which neutrophil retentionin a filter increases substantially,7' which isrelevant to neutrophil retention in similarlysized pulmonary capillary segments.Monocytes and neutrophils have similardiameters and are retained when radiola-belled and reinjected in the rabbit lung to agreater extent than the smaller diameter lym-phocytes.78 These data suggest that cell sizeinfluences leucocyte retention in the lungs.Monocytes are retained to a greater extentthan neutrophils, however, despite being ofsimilar diameter,7879 indicating that factorsother than cell size are also important indetermining the transit of leucocytes in thelungs. Downey and coworkers80 andDoerschuk et al 56 were both able to show thatwhen neutrophils were artificially stiffened byfixation in glutaraldehyde substantially morewere retained in rabbit lungs. Using both afiltration technique and the "cell poker" tomeasure deformability in vitro, Worthen eta181 also showed that populations of neu-trophils with differing deformability had dif-fering retention rates in the lungs of rabbits invivo: the more deformable the cells, the fewerwere retained in the lungs.

Neutrophil adhesion, at least that mediatedby the CD11/18 integrin (see below), doesnot have a major role in the retention ofunstimulated neutrophils in micropore mem-branes. In support of this hypothesis, differ-entiated cells of the myelomonocytic HL60cell line, which express CD11/18, wereretained less than undifferentiated cells,which do express this antigen.82 These differ-ences in retention may be accounted for bydifferences in cell size and stiffness.Moreover, the decreased cell deformabilitythat accompanies activation of neutrophilswith N-formyl-methionyl-leucyl-phenylalanine(FMLP) did not change when HL60 cellswere treated with the monoclonal antibody603 to the common beta chain of theCD 1/18 integrin.8' The decreased deforma-bility induced by cell activation appears to bedue to assembly of the cytoskeleton, particu-larly the microfilaments.8384 The decreased

0.35 r

Figure 3 The significantlinear relationship seen innormal subjects between theinitial pressure gradientgenerated when neutrophilsare filtered in vitro andfirst pass sequestrationwhen autologousneutrophils are filteredsimultaneously in vivo bythe pulmonary vascularbed (r = 09, p = O001).Reproducedfrom Selby etal'7 by courtesy of theAmerican PhysiologicalSociety.

c

0._

4_4-tn

0)r

U0)Cl)

._

mE

0

:,'

0301

0251

0-20 [

0-151

0 10

005

00

8

0

0 5 10Pressure gradient (cm H20 min-')

cell deformability induced by FMLP wascompletely abolished by treating HL60 cellswith cytochalasin D, which disrupts actin.82Moreover, neutrophils activated with FMLPand treated with cytochalasin D were notretained in the rabbit lung more than cellstreated with FMLP alone.8' These studiesconfirm the role of cell deformability as amechanism of neutrophil retention in the pul-monary microcirculation in vivo in animals.The importance of the cell's cytoskeletalassembly in the retention of neutrophils in thepulmonary microcirculation is emphasised bystudies of cells of the HL60 line, which differ-entiate towards granulocytes.8' During differ-entiation these cells change the organisationof their actin in response to stimulation withFMLP, though they are unable to increasetheir adhesiveness. Disruption of the cellmicrofilament organisation with cytochalasinD abolished the increase in retention in vitroof both undifferentiated and FMLP stimulat-ed differentiated cells, confirming the impor-tant influence of microfilament organisationon the rheological properties of the neu-trophil.82

These in vitro studies of neutrophil filtra-tion through capillary sized pores haverecently been applied to the problem ofneutrophil sequestration in vivo in man.We showed a significant correlation betweenthe deformability of neutrophils in vitro andtheir sequestration during their first transit inthe pulmonary vasculature after reinjection innormal subjects57 (fig 3). This study confirmsthe hypothesis that cell deformability is animportant determinant of the normal seques-tration of neutrophils in the lungs due to thegeometric constraints imposed on neutrophilsin transit in the pulmonary capillary bed.Such increased intravascular sequestration ofneutrophils is enhanced in inflammation ofthe lung in animal models,56 and by directvideomicroscopy has been seen to occur inthe capillary bed rather than in the postcapil-lary venules,85 in contrast with the systemiccirculation.

Increased capillary sequestration of neu-trophils occurs whether the inflammatorystimulus is administered via the airways or viathe blood.8687 Using an animal model ofneutrophil sequestration,88 which has beenverified by studies in man,89 we have alsoshown that activated neutrophils are sequesteredin normal lungs more than quiescent neu-trophils in the inflamed lung. This increasedsequestration is associated with a decrease incell deformability90 and with increased migra-tion of neutrophils into the air spaces.90Moreover, sequestration of activated neu-Lrophils in itself may be associated with bothepithelial and endothelial injury.86As might be expected from the preceding

discussion, conditions where neutrophildeformability decreases in man should beassociated with increased neutrophil seques-tration in the lungs. This has been confirmedin patients with chronic obstructive lung dis-ease during acute exacerbations, who havesignificantly greater retention of radiolabelled

0-00 o

84



/T


nloaded from


Neutrophil traffic in the lungs: role ofhaemodynamics, cell adhesion, and deformability

NEUTROPHIL

L-Selectin CD1 1 a/CD18 CD11b/CD18 CD1lc/CD18(LECAM-1) (LFA-1) (Mac-1) (pl 5095)

P-Selectin(GMP140,CD62)

ICAM-1 ICAM-2(CD54)

E-Selectin(ELAM-1)

HF

ENDOTHELIUM

Figure 4 Adhesion molecules, currently considered to be important in neutrophendothelium interactions. The double solid lines are the respective plasma membblack boxes representing the transmembrane domains. Cytoplasmic domains are

represented by curved lines. The common (smaUer) CD18 chain is represented bportion of the arrow heads. Current (and previous or alternative) nomenclatureindicated.

neutrophils than clinically stable patients, andthis increased retention is associated with adecrease in cell deformability measured invitro.9' These changes resolved when thepatients became clinically stable.57 Cigarettesmoking is also associated with increasedsequestration of neutrophils in the lungs."Exposure of neutrophils to cigarette smoke invitro results in decreased deformability asso-ciated with polymerisation of actin.9' We havealso recently shown that cigarette smoking isassociated with an acute decrease in thedeformability of neutrophils in arterial bloodin vivo,93 which is further confirmation of theimportant influence of changes in neutrophildeformability on cell sequestration in the pul-monary vasculature. Preliminary data alsoindicate that increased neutrophil sequestra-tion during smoking results in increased elas-tase concentrations in rapidly sampled arterialblood,94 which may increase the protease bur-den in the lungs and may be relevant to thepathogenesis of emphysema.57 As describedabove, activation of neutrophils is associatednot only with a rapid decrease in theirdeformability but also with changes in theiradhesion to each other and to other cells.Decreased deformability seems likely to bethe initiating event producing neutrophildelay or entrapment in the pulmonary micro-circulation and allowing adhesive interactionbetween neutrophils and endothelial cells toproceed thereafter.

Cell adhesionThree principal families of adhesion receptors

have been identified in leucocyte-endothelial

cell adhesion: the integrins, those belongingto the immunoglobulin supergene family, andthe vascular selectins. Recent reviews have

e considered adhesion of leucocytes, includingneutrophils, to both endothelial and epithelialcells95-97 and their relevance to neutrophilemigration.98

Integrins are transmembrane heterodimericglycoproteins comprising large a and smaller,B chains. Among these are the leucocyte celladhesion molecules (leuCAMs), which con-sist of a common fl2 chain; the CD18 antigen,associated with one of three a chains: theCD1 1 (a-c) antigens.9699 CD18/CD 1 a andCD18/CDllb are constitutively present onthe surface of human peripheral blood neu-trophils.9699 When cells are activated augmen-tation of integrins occurs both by recruitmentfrom intracellular sources, particularly fromspecific granules,100 and through conforma-

e tional changes, with the revelation of neoepi-topes'0' and phosphorylation.96 These changesare usually associated with increased celladhesivity. There is, however, a dichotomybetween increased CD18/CD 11 expressionand increased cell adherence.'02 Henceimmunostaining of cell surface integrins doesnot provide a precise indicator of the func-tional state of cell adhesiveness. This sectionof the review focuses on intravascular neu-trophil adhesion and we have therefore notdiscussed the ,B or the fi, families of

I integrins.98L LeuCAM-ligand interaction initiates aI series of signal transduction mechanisms that

are only now beginning to be unravelled; theyinclude protein phosphorylation and proteinkinase C activation with cytoskeletal confor-mational changes.'03The intravascular ligands for the leuCAM

integrins CD1la/CD18 and CD1lb/CD18are single chain glycoproteins of theimmunoglobulin supergene family known asintercellular adhesion molecules (ICAMs).Both ICAM-1'04 and ICAM-2'05 are constitu-tively present on endothelium (fig 4).ICAM-1 is also present on the basal surfaceof bronchial epithelial cells'06 and probablythe luminal surface of type II alveolar pneu-mocytes.Of the vascular selectins, L-selectin is pre-

sent constitutively on circulating unactivatedneutrophils'07 and a further two, E-selectin'08and P-selectin,'09 are inducible on endothelialsurfaces. In the systemic circulation neu-trophil L-selectin interacts probably withP-selectin and an additional uncertainendothelial ligand, which mediate low leveladhesion and the rolling of neutrophils alongthe endothelium."0 Thereby shear stresses arereduced, which allows neutrophils to developstronger adhesion to endothelium mediatedby CD18/CD1I-ICAM or E-selectin.98 Thisbrings the neutrophil to a halt in preparationfor subsequent diapedesis, whereupon L-selectin appears to be proteolytically cleavedfrom the neutrophil surface."' There are fewdata on the role of L-selectin in the pul-monary circulation, but it would be facile andprobably erroneous to postulate mechanisms

i Ir

85



/T


nloaded from


MacNee, Selby

similar to those in the systemic circulation.Preliminary data suggest that L-selectin is notnecessary for sequestration of neutrophils inthe lung in response to the infusion of activat-ed complement in the rabbit."2 Neither issuch sequestration associated with loss of L-selectin from the neutrophils."'What of the other side of the intravascular

equation, the endothelium? Treatment ofendothelium with thrombin mobilises P-selectin within minutes from intracellularWeibel-Palade bodies.' '4 This selectin canbind normal neutrophils,"° probably via L-selectin, and initiate leucocyte rolling. Its rolein the pulmonary circulation is not clear.

Activation of endothelium by proinflam-matory mediators, such as the cytokines IL-1and TNF, or by lipopolysaccharide inducesthe sequential expression of E-selectin'08 fol-lowed by ICAMs,'06 which is blocked byinhibitors of protein synthesis. This novelexpression of E-selectin peaks around fourhours after stimulation,'08 in contrast with theup regulation of vascular ICAM-1, which ismaximal at around 24hrs.'06 The ligand for E-selectin on the neutrophil has yet to be con-clusively characterised but is probablysialated Lewis X.' '5The crucial role of neutrophil adherence,

particularly the functioning of leuCAMs invivo in man, is graphically illustrated by theleucocyte adhesion molecule deficiency syn-drome, which results from a congenitalinability to express functional CD18/CD 11.116Neutrophils in these patients display impairedadherence in vitro, and an increased intravas-cular half life in vivo."7 These neutrophils,however, sequester normally within the pul-monary vasculature and are released normallyduring vigorous exercise."7 The syndrome isfatal, death being caused usually by over-whelming mucous membrane and skin infec-tion, but rarely by pneumonia."6CD18/CD 11 does not appear to be neces-

sary for the sequestration of normal neu-trophils within a normal pulmonary vascularbed."18 Thus the initial excess of neutrophilswithin the capillaries of the normal lung mayresult from the adverse geometrical con-straints and deformability characteristics con-sidered previously, which impede the transitof cells.

Neutrophil migration, however, but notintravascular sequestration within the lungs,in response to the inflammation induced byintrabronchial phorbol ester was inhibited bya monoclonal antibody to CD18. In contrast,intravascular sequestration and migration inthe lungs in response to pneumococciinstilled intrabronchially was not abolished,which was not the case when the same stimu-lus was applied to the skin."9 Thus the mech-anisms mediating neutrophil migration notonly are stimulus specific but also differ in thepulmonary and systemic circulations. Thismay be due to differences in endothelialresponsiveness or to the proximity of othereffector cells, such as epithelial and phagocyt-ic cells, which enhance recruitment of neu-trophils within the lungs. It would indeed be

unusual for CD18 alone to be a central pivotin such mechanisms. The rarity of pneu-monia in patients lacking functionalCD 18/CD 1 1 in the leucocyte adhesion mole-cule deficiency syndrome confirms alterna-tive, CD 18-independent, pathways foreffective neutrophil recruitment.Thus treatment with monoclonal antibod-

ies directed against CD18 may be of limitedbenefit if not detrimental. In the few studiesin man investigating this antibody in patientsundergoing bone marrow transplantation lit-tle consistent success has been claimed.'20 121There may be more hope with the develop-ment of monoclonal antibodies that can iden-tify epitopes of activated CD 18/CD 1 1.Blocking endothelial ligands is an alternativeapproach. Monoclonal antibodies to ICAM-1protect renal allografts in primate recipients'22and have attenuated eosinophil infiltrationand airway hyperresponsiveness in a primatemodel of asthma.'23 Short peptides have beenfound to block ICAM-1 function in vitro'24and may be more attractive clinically.

Such treatment, however, may be effectivefor only a finite time window of opportunity."3Future study of these factors that controlneutrophil traffic in the lungs will provideexciting insights into the mechanisms of lunginflammation and injury and indicate novelavenues of therapeutic intervention.

1 Hogg JC. Neutrophil kinetics and lung injury. Physiol Rev1987;67: 1249-95.

2 MacNee W, Selby C. Neutrophil kinetics in the lungs.Clin Sci 1990;79:97-107.

3 Williams JH. Indium labelled neutrophils for detectinglung injury: a perspective. Chest 199 i;99: 1248-51.

4 Tate RM, Repine JE. Neutrophils and the adult respirato-ry distress syndrome. Am Rev Respir Dis 1983;128:552-9.

5 Neiwoehner DE. Cigarette smoking, lung inflammationand the development of emphysema. J Lab Clin Med1988;111:15-27.

6 Cochrane CG. Cellular injury by oxidants. Am J Med1991;91(3C):23-30S.

7 Janoff A. Elastases and emphysema. Current assessmentof the protease-antiprotease hypothesis. Am Rev RespirDis 1985;132:417-33.

8 Athens JW, Haab OP, Raab SO, Mauer AM,Ashenbrucker H, Cartwright GE, et al. The total bloodcirculating and of marginal granulocyte pools and thegranulocyte turnover rate in normal subjects. Y7 ClinInvest 1961;40:989-95.

9 Malpighi M. De Pulmonivus (1661), translated by J.Young. Proc R Soc Med 1930;23: I.

10 Weibel ER. Morphometry of the human lung. Berlin:Springer, 1963:78-82.

11 Fung YC, Sobin SS. Theory of sheet flow in lung alveoli.YAppl Physiol 1969;26:472-88.

12 Sobin SS, Tremer HM, Fung YC. Morphometric basis ofsheet-flow concept of the pulmonary alveolar microcir-culation in the cat. Circ Res 1970;26:397-414.

13 Guntheroth WG, Luchtel DL, Kawabori I. Pulmonarymicrocirculation: tubules rather than sheet and post. YAppl Physiol 1982;53:510-5.

14 Weibel ER. Lung cell biology. Handbook ofphysiology. Therespiratory system. Bethesda: American PhysiologicalSociety, 1984;Chap 2.p47-91.

15 Schmid-Schonbein GW, Shih YY, Chien S. Morphometryof human leukocytes. Blood 1985;56:866-75.

16 Beyers N, Doerschuk CM, Coxson H, Hogg JC.Neutrophil volume and capillary segment diameter inthe rabbit, the dog and the human [abstract]. Am RevRespir Dis 1989;139:A297.

17 West JB, Dollery CT, Naimark A. Distribution of bloodflow in isolated lung; relation to vascular and alveolarpressures. IApplPhysiol 1964;19:713-24.

18 Glazier JB, Hughes JMB, Maloney JE, West JB.Measurements of capillary dimensions and blood vol-ume in rapidly frozen lungs. .7 Appl Physiol 1969;26:65-76.

19 Schmid-Schonbein GW, Usami S, Skalak R, Chien S.The interaction of leukocytes and erythrocytes in capil-lary and postcapillary vessels. Microvasc Res 1980;

86



/T


nloaded from



19:45-70.20 Ernst E, Hammerschmidt DE, Bagge U, Matrai A,

Dormandy JA. Leukocytes and the risk of ischaemic dis-eases. JAMA 1987;257:2318-24.

2.1 Nagasaka Y, Bhattacharya J, Gropper MA, Staub NC.Micropuncture measurements of lung microvascularpressure profile during hypoxia in cats. Circ Res1984;54:90-5.

22 Wasserman K, Butler J, Vankessel A. Factors affecting thepulmonary capillary blood flow pulse in man. Jf ApplPhysiol 1966;21:890-900.

23 Folkow B, Niel E. The circulation. Oxford: OxfordUniversity Press, 1971:37.

24 Cartwright GE, Athens JW, Wintrobe MM. The kineticsof granulopoiesis in normal man. Blood 1964;24:780-803.

25 Athens JW, Raab SO, Haab OP, Mauer AM,Ashenbrucker H, Cartwright GE, et al. Leukokineticstudies. III. The distribution of granulocytes in theblood of normal individuals. J Clin Invest 1961;40:159-64.

26 Savill JS, Wylie AH, Henson JE, Walport MJ, HensonPM, Haslett C. Macrophage phagocytosis of ageingneutrophils in inflammation. Programmed cell death inthe neutrophil leads to recognition by macrophages.J Clin Invest 1989;83:865-75.

27 Bierman HR, Kelly KH, Cordes FL, Byron RL,Polheumus J, Rappoport S. The release of leukocytesand platelets from the pulmonary circulation by epi-nephrine. Blood 1952;7:683-92.

28 Cohnheim J. Lectures on general pathology. Vol 1, 2nd ed.[translated from 2nd German edition]. London: NewSydenham Society. 1889:100-12.

29 Atherton A, Born GVR. Relationship between velocity ofrolling granulocytes and that of blood flow in venules. J7Physiol Lond. 1973;233: 157-65.

30 Haslett C, Guthrie LA, Kopaniak MM, Johnston Jr. RB,Henson PM. Modulation of multiple neutrophil func-tions by preparative methods or trace concentrations ofbacterial lipopolysaccharide. Am _J Pathol 1985;119:101-10.

31 McAfee JG, Subramanian G, Gagne G. Technique ofleukocyte harvesting and labelling: problems and per-spectives. Semin Nucl Med 1984;14:83-106.

32 Saverymuttu SH, Peters AM, Danpure HJ, Reavy HJ,Osman S, Lavender JP. Lung transit of "'In-labelledgranulocytes. Relationship to labelling techniques.ScandJ Haematol 1983;30:151-60.

33 Peters AM, Saverymuttu SH, Bell RN, Lavender JP.Quantification of the distribution of the marginatinggranulocyte pool in man. Scand _J Haematol 1985;34:111-20.

34 Chien S. Role of blood cells in microcirculatory regula-tion. Microvasc Res 1985;29:129-51.

35 Markos J, Hooper RO, Kavanagh-Gray D, Wiggs BR,Hogg JC. Effect of raised alveolar pressure on leukocyteretention in the human lung. J Appl Physiol 1990;69:214-21.

36 Doerschuk CM, Allard MF, Martin BA, MacKenzie A,Autor AP, Hogg JC. Marginating pool of neutrophils inrabbit lungs. JAppl Physiol 1987;63:1806-15.

37 Lien DC, Wagner WW, Capen RL, Haslett C, HansonWL, Hofmeister SE, et al. Physiological neutrophilsequestration in the lung: visual evidence for localisationin capillaries. J Appl Physiol 1987;62:1236-43.

38 Martin BA, Wright JL, Thommasen H, Hogg JC. Effectof pulmonary blood flow on the exchange between thecirculating and marginating pool of polymorphonuclearleukocytes in dog lungs. J Clin Invest 1982;69: 1277-85.

39 Thommasen HV, Martin BA, Wiggs BR, Quiroga M,Baile EM, Hogg JC. Effect of pulmonary blood flow onleukocyte uptake and release by dog lung. J Appl Physiol1984;56:966-74.

40 Doerschuk CM, Allard MF, Brumwell ML, Hogg JC.Effect of epinephrine on neutrophil kinetics in rabbitlungs. JAppl Physiol 1988;65:401-7.

41 Martin BA, Wiggs BR, Lee S, Hogg JC. Regional differ-ences in neutrophil margination in dog lungs. _J ApplPhysiol 1987;63:1253-61.

42 Muir AL, Cruz M, Martin BA, Thommasen H, BelzbergA, Hogg JC. Leukocyte kinetics in the human lung: roleof exercise and catecholamines. J Appl Physiol 1984;57:711-9.

43 Ohgami M, Doerschuk CM, English D, Dodek PM,Hogg JC. Kinetics of radio-labelled neutrophils inswine. J Appl Physiol 1989;66:1881-5.

44 MacNee W, Wiggs B, Belzberg AS, Hogg JC. The effectof cigarette smoking on neutrophil kinetics in hurmanlungs. N Engl_JMed 1989;321:924-8.

45 Brumwell ML, MacNee W, Doerschuk CM, Wiggs BR,Hogg JC. Neutrophil kinetics in normal and emphyse-matous regions of human lung. Ann NY Acad Sci199 1;624:30-9.

46 Perlo S, Jalowayski AA, Durant DM, West JB.Distribution of red and white blood cells in alveolarwalls. J ApplPhysiol 1975;38:117-24.

47 Chien S. White blood cell rheology. In: Lowe GD, ed.Clinical blood rheology. Boca Raton: CRC Press,

1988:87-110.48 Staub NC, Schultz EL. Pulmonary capillary length in dog,

cat and rabbit. RespirPhysiol 1968;5:371-8.49 Capen RL, Latham LP, Wagner WW. Comparison of

direct and indirect measurements of pulmonary capil-lary transit times. _ Appl Physiol 1987;62: 1150-4.

50 MacNee W, Martin BA, Wiggs BR, Belzberg A, Hogg JC.Regional pulmonary transit times in man. J Appl Physiol1989;66:844-50.

51 Hogg JC, Martin BA, Lee S, McLean T. Regional differ-ences in erythrocyte transit in normal lungs. J ApplPhysiol 1985;59:1266-71.

52 Hogg JC, McLean T, Martin BA, Wiggs B. Erythrocytetransit and neutrophil concentration in the dog lung. J

Appl Physiol 1988;65:1217-25.53 Donnelly SC, Haslett C. Cellular mechanisms of acute

lung injury: implications for future treatment of theadult respiratory distress syndrome. Thorax 1992;47:260-3.

54 Thomassen HV, Boyko WJ, Russell JA, Hogg JC.Transient leukopenia associated with adult respiratorydistress syndrome. Lancet 1984;i:809-12.

55 Worthen GS, Smedley SLA, Tonnesen MG, Ellis D,Voelkel NF, Reeves JT, et al. Effects of shear stress onadhesive interaction between neutrophils and culturedepithelial cells. J Appl Physiol 1987;63:2031-41.

56 Doerschuk CM, Allard MF, Hogg JC. Neutrophil kineticsin rabbits during infusion of zymosan-activated plasma._JAppl Physiol 1989;67:88-95.

57 Selby C, Drost E, Wraith PK, MacNee W. In vivoneutrophil sequestration within the lungs of man isdetermined by in vitro "filterability." J Appl Physiol1991;71:1996-2003.

58 Lien DC, Worthen GS, Capen RL, Hanson WL,Checkley LL, Janke SJ, et al. Neutrophil kinetics in thepulmonary microcirculation. Effects of pressure andflow in the dependent lung. Am Rev Respir Dis1990;141:953-9.

59 Peters AM, Allsop P, Stuttle AWJ, Arnot RN, William M,Hall GM. Granulocyte margination in the human lungand its response to strenuous exercise. Clin Sci 1992;82:237-44.

60 Lowe GDO. Blood rheology in vitro and in vivo. Bailliere'sClinical Haematology 1987;1:597-636.

61 Bagge U. Leukocytes and capillary perfusion in shock. In:Meiselmann HG, Lichtman MA, LaCelle PL, eds.W'hite cell mechanics: basic science and clinical aspects. NewYork: Liss, 1984:285-94.

62 Schmid-Schonbein GA, Sung KLP, Tozeren H, Skalak R,Chein S. Passive mechanical properties of human leuko-cytes. BiophysJ7 1981;34:243-7.

63 Schmid-Schonbein GW, Skalak R, Sung KLP, Chien S.Human leukocytes in the active state. In: Bagge U, BornGV, Gaehtgens P, eds. White blood cells, morphology andrheology as related to function. The Hague: MartinusNijhoff, 1984:21-31.

64 Warshawski FJ, Sibbald WJ, Driedger AA, Cheung H.Abnormal neutrophil pulmonary interaction in the adultrespiratory distress syndrome. Qualitative and quantita-tive assessment of pulmonary neutrophil kinetics inhumans with in vivo "'-indium neutrophil scintigraphy.Am Rev Respir Dis 1986;133:797-804.

65 Chien S, Schmid-Schonbein GW, Sung KLP, SchmalzerEA, Skalak R. Viscoelastic properties of leukocytes. In:Meiselmann HG, Lichtman MA, LaCelle PL, eds.White cell mechanics: basic science and clinical aspects. NewYork: Lss, 1984:19-51.

66 Schmid-Schonbein GW, Sung KLP, Tozeren H. SkalakR, Chien S. Passive mechanical properties of humanleukocytes. Biophys J 1981;36:243-56.

67 Evans EA, Kukan B. Passive material behaviour of granu-locytes based on large deformation and recovery afterdeformation tests. Blood 1984;64:1028-35.

68 Chien S, Schmalzer EA, Lee MML, Impelluso T, SkalakR. Role of white blood cells in filtration of blood sus-pensions. Biorheology 1983;20: 11-5.

69 Reinhart WH. The time course of a filtration test as amodel for microvascular plugging by white cells andhardened red cells. Microvasc Res 1987;34:1-12.

70 Skalak R, Impelluso T, Schmalzer EA, Chien S.Theoretical modelling of filtration of blood cell suspen-sions. Biorheology 1983;20:41-56.

71 Reinhart WH, Usami S, Schmalzer EA, Lee MM, ChienS, Evaluation of red blood cell filterability test: influ-ences of pore size, hematocrit level and flow rate. J7 LabClin Med 1984;104:501-16.

72 Nash GB, Jones JG, Mikita J, Dormandy JA. Methodsand theory for analysis of flow of white cell subpopula-tions through micropore filters. Br JfHaematol 1988;70:165-70.

73 Downey GP, Worthen GS. Neutrophil retention in modelcapillaries: deformability, geometry and hydrodynamicforces. I Appl Physiol 1988;65:1861-71.

74 Evans EA. New membrane concept applied to the analysisof fluid-shear and micro-pipette-deformed red bloodcells. BiophysJ 1973;13:941-54.

75 Peterson NO, McConnaughey MM, Elson EL.Dependence of locally measured cellular deformability

87



/T


nloaded from


MacNee, Selby

on position on the cell, temperature, and cytochalasinB. Proc NadAcad Sci 1982;79:5327-31.

76 Dailey B, Halak GI, Cell poking determination of the elas-tic area compressibility modulus of the red cell mem-brane. BiophysyJ 1984;45:671-82.

77 Lennie SE, Low GDO, Barberel JC, Forbes CD, FouldsWS. Filterability of white blood cell subpopulations,separated by an improved method. Clin Haemorheol1987;7:811-6.

78 Doerschuk CM, Downey GP, Doherty DE, English D,Gie RP, Ohgami M, et al. Leukocyte and platelet mar-gination within microvasculature of rabbit lungs. Y ApplPhysiol 1990;68: 1956-61.

79 Doherty DE, Downey GP, Worthen GS, Haslett C,Henson PM. Monocyte retention and migration in pul-monary inflammation: requirement for neutrophils. LabInvest 1988;59:200-13.

80 Downey GP, Doherty DE, Schwab B, Elson EL, HensonPM, Worthen GS. Retention of leukocytes in capillaries:role of cell size and deformability. Y Appl Physiol1990;69: 1767-78.

81 Worthen GS, Schwab B, Elson EL, Downey GP.Mechanics of stimulated neutrophils: cell stiffeninginduces retention in capillaries. Science 1989;245:183-6.

82 Erzurum SC, Kus ML, Bohse C, Elson EL, Worthen GS.Mechanical properties of HL60 cells: role of stimulationand differentiation in retention in capillary-sized pores.AmJ Respir Cell Mol Biol 1991;5:230-4 1.

83 Omann GM, Allen RA, Bokoch GM, Painter RG,Traynor AE, Sklar LA. Signal transduction andcytoskeletal activation in the neutrophil. Physiol Rev1987;67:285-322.

84 Elson EL. Cellular mechanics as an indicator of cytoskele-tal structure and function. Annu Rev Biophys BiophysChem 1988;17:397-420.

85 Lien DC, Henson PM, Capen RL, Henson JE, HansonWL, Wagner WW, et al. Neutrophil kinetics in the pul-monary microcirculation during acute inflammation.Lab Invest 1991;65:145-59.

86 Gie RP, Doerschuk CM, English D, Coxon HO, HoggJC. Neutrophil associated lung injury after an infusionof activated plasma. JfAppl Physiol 1991 ;70:2471-8.

87 Peters MJ, Breslin ABX, Kemp AS, Chu J, Berend N.Haematological effects of inhalation of N-formyl-methionyl-leucyl-phenyalanine in man. Thorax 1992;47:284-7.

88 Brown GM, Drost E, Selby C, Donaldson K, MacNee W.Neutrophil sequestration in rat lungs. Ann NYAcad Sci1991;624:316-7.

89 Selby C, Drost E, O'Neil S, Gillooly M, Lamb D,Cameron E, et al. Does emphysema affect neutrophiltraffic through the lungs? [abstract]. Am Rev Respir Dis1991;143:A790.

90 Brown GM, Drost E, Donaldson K, MacNee W.Sequestration of neutrophils in rat lungs [abstract]. AmRev Respir Dis 1991;143:A230.

91 Selby C, Drost E, Lannan S, Wraith PK, MacNee W.Neutrophil retention in the lungs of patients with chron-ic obstructive pulmonary disease. Am Rev Respir Dis1991;143:1359-64.

92 Drost EM, Selby C, Lannan S, Lowe GDO, MacNee W.Changes in neutrophil deformability following in vitrosmoke exposure mechanism and protection. Am J RespCell Mol Biol 1992;6:287-95.

93 Drost EM, Selby C, Bridgeman M, Marsden M, MacNeeW. Oxidant effects of smoke on components of wholeblood [abstract] . Am Rev Respir Dis 1992;145:A577.

94 MacNee W, Bell D, Dawes J, Muir AL, Hogg JC.Changes in the levels of human neutrophil elastase andproducts of lipid peroxidation during smoking[abstract]. Am Rev Respir Dis 1989;139:A574.

95 Patarroyo M, Makgoba MW. Leucocyte adhesion to cells:molecular basis, physiological relevance and abnormali-ties. Scand J Immunol 1989;30: 129-64.

96 Albelda SM. Endothelial and epithelial cell adhesion mol-ecules. Am J Respir Cell Mol Biol 1991;4: 195-203.

97 Ley K. Leukocyte adhesion-molecular basis and physio-logical consequences. Clin Hemorheol 1992;12:93-108.

98 Harlan JM, Schwartz BR, Wallis WJ, Pohlman TH. Therole of neutrophil membrane proteins in neutrophil emi-gration. In: Movat HZ ed. Leukocyte emigration and itssequel. Basel: Karger, 1986:94-104.

99 Cobbold S, Hale G, Waldmann H. Nonlineage, LFA-1family and leucocyte common antigens: new and previ-ously defined cluster. In: McMichael AJ, ed. Leucocytetping. III. White cell differentiation antigens. Oxford:Oxford University Press, 1987:788-803.

100 Todd RF, Arnaout MA, Rosin RE, Crowley CA, PetersWA, Babior BM. Subcellular localisation of the largesubunit of Mol (Mol alpha; formerly gp 110), a surfaceglycoprotrein associated with neutrophil adhesion. JClin Invest 1984;74:1280-90.

101 Altieri DC, Edgington TS. A monoclonal antibody react-

ing with distinct adhesion molecules defines a transitionin the functional state of the receptor CDllb/CD18

(Mac- 1). J Immunol 1988;141:2656-60.102 Philips MR, Buyon JP, Winchester R, Weissmann G,

Abramson SB. Up-regulation of the iC3b receptor(CR3) is neither necessary nor sufficient to promoteneutrophil aggregation. J Clin Invest 1988;82:495-50 1.

103 Lew DP. Receptor signalling and intracellular calcium inneutrophil activation. EurJ Clin Invest 1989;19:338-46.

104 Rothlein R, Dustin MI, Marlin SD, Springer TA. Ahuman intercellular adhesion molecule (ICAM-1) dis-tinct from LFA-1.3'Immunol 1986;137:1270-4.

105 Staunton DE, Dustin ML, Springer TA. Functionalcloning of ICAM-2, a cell adhesion ligand for LFA-1homologous to ICAM-1. Nature 1989;339:61-4.

106 Dustin ML, Rothlein R, Bhan AK, Dinarello CA,Springer TA. Induction by IL-1 and interferon, tissuedistribution, biochemistry and function of a naturaladherence molecule (ICAM-1). J Immunol 1986;137:245-54.

107 Jutila KL, Kishimoto TK, Butcher EC. Regulation andlectin activity of the human neutrophil peripheral lymphnode homing receptor. Blood 1990;76: 178-83.

108 Bevilacqua MP, Stengelin S, Gibmore MA, Seed B.Endothelial leukocyte adhesion molecule 1: an induciblereceptor for neutrophils related to complement regula-tory proteins and lectins. Science 1989;243:1160-4.

109 Bonfanti R, Furie BC, Wagner DD. PADGEM(GMP140) is a component of Weibel-Palade bodies ofhuman endothelial cells. Blood 1989;73:1 109-12.

110 Lawrence MB, Springer TA. Leukocytes roll on aselectin at physiological flow rates: distinction from andprerequisite for adhesion through integrins. Cell 1991;65:859-73.

111 Kishimoto TK, Jutila MA, Berg EL, Butcher EC.Neutrophil Mac-i and MEL-14 adhesion proteinsinversely regulated by chemotactic factors. Science1989;245:1238-41.

112 Doyle NA, Quinlan WM, Graham L, Doerschuk CM.The role of LECAM-1 in neutrophil sequestrationinduced by infusion of complement fragments[abstract]. Am Rev Respir Dis 1992;145:A187.

113 Quinlan WM, Doyle NA, Graham L, Doerschuk CM.Neutrophil distribution and LECAM-1 expression dur-ing and after infusion of complement fragments[abstract]. Am Rev RespirDis 1992;145:A187.

114 Hattori R, Hamilton KK, Fugate RD, McEver RP, SimsPJ. Stimulated secretion of endothelial von Willebrandfactor is accompanied by rapid redistribution to the cellsurface of the intracellular granule membrane proteinGMP-140._ Biol Chem 1989;264:7768-71.

115 Philips ML, Nudelman E, Gaeta FCA, Perez M, SinghalAK, Hakomori S-I, Paulson JC. ELAM-1 mediates celladhesion by recognition of a carbohydrate ligand sialyl-Lex. Science 1990;250:1130-2.

116 Anderson DC, Springer TA. Leukocyte adhesion defi-ciency: an inherited defect in the Mac-i, LFA-1 andpl 50-95 glycoproteins. Annu Rev Med 1987;38:175-94.

117 Davies KA, Toothill VJ, Savill J, Hotchin N, Peters AM,Pearson JD, et al. A 19 year old man with leucocyteadhesion deficiency. In vitro and in vivo study of leuco-cyte function. Clin Exp Immunol 1991;84:223-31.

118 Yoder MC, Checkley LL, Giger U, Hanson WL, KirkKR, Capen RL, et al. Pulmonary microcirculatory kinet-ics of neutrophils deficient in leukocyte adhesion-pro-moting glycoproteins. J Appl Physiol 1990;69:207-13.

119 Doerschuk CM, Winn RK, Coxson HO, Harlan JM.CD 18-dependent and -independent mechanisms ofneutrophil emigration in the pulmonary and systemicmicrocirculation of rabbits. 7 Immunol 1990;144:2327-33.

120 Fischer A, Friedrich W, Fasth A, Blanche S, le Deist F,Girault D, et al. Reduction of graft failure by a mono-

clonal antibody (anti-LFA-l CD lla) after HLA non-

identical bone marrow transplantation in children withimmunodeficiencies, osteopetrosis and Fanconi'sanemia: a European Group for Immunodeficiency/European Group for Bone Marrow Transplantationreport. Blood 1991;77:249-56.

121 Baume D, Kuentz M, Pico JL, Beaufean F, CordonnierC, Vernant JP, et al. Failure of a CD18/anti-LFAImonoclonal antibody infusion to prevent graft rejectionin leukemic patients receiving T-depleted allogeneicbone marrow transplantation. Transplantation 1989;47:471-4.

122 Cosimi AB, Cont D, Delmonico FL, Preffer FI, Wee SL,Rothlein R, et al. In vivo effects of monoclonal antibodyto ICAM-1 (CD54) in nonhuman primates with renalallografts. J Immunol 1990;144:4604-12.

123 Wegner CD, Gundel RH, Reilly P, Haynes N, Letts LG,Rothlein R. Intercellular adhesion molecule-i (ICAM-1) in the pathogenesis of asthma. Science 1990;247:456-9.

124 Fecondo IV, Kent SB, Boyd AW. Inhibition of intercel-lular adhesion molecule 1-dependent biological activi-ties by a synthetic peptide analog. Proc Natl Acad SciUSA 1991;88:2879-82.

88



/T


nloaded from


the neutrophil lungs: haemodynamics, cell andtion (fig 2). the regional retention of radiolabelled...

Documents