Surface-activated Bovine Platelets Do NotSpread, They Unfold
Lynette H. Grouse,* Gundu H. R. Rao,* DouglasJ. Weiss,t Victor Perman,t and James G. WhitetFrom the Departments ofLaboratory Medicine andPathology,* Pediatrics,t and Veterinary Pathobiology,jthe University ofMinnesota Health Sciences Center,Minneapolis, Minnesota
Thepresent study has examined the response ofbo-vine platelets to surface activation and comparedit to the reaction ofhuman cells. Human plateletsreact to surfaces by losing their discoid shape, ex-tendingpseudopods, converting to dendriticforms,and finally, spreading into thin films resemblingpancakes. Bovine platelets do not spread, they un-fold. Surface activation causes them to transformfrom discs to irregular, flattened shapes resemblingdendritic platelets, but they are unable to fill inspaces between pseudopods, a step required forspreading. Bovine platelets lack the surface-con-nected open canalicular system (OCS), whichserves as a reservoir ofmembranefor humanplate-let spreading. Its absence may be the majorfactorin thefailure ofbovineplatelet spreading, but thereare other possible factors. Circumferential micro-tubules are more resistant to disassembly in sur-face-activated bovine than human cells, and theirstability as rings or fractured bundles may limitspreading. Actinfilament assembly is similar in hu-man and bovine platelets, but the organization isdifferent. Humanplateletsform aperipheral weaveofactin that expands the membrane between pseu-dopods. A peripheral weave does notform in sur-face-activated bovine platelets. The absence of theOCSand differences in cytoskeletal organization inbovine platelets may also affect spreading of thesurface membrane. Fibrinogen-gold (Fgn-Au)probes added to spread human platelet movefrompseudopods and the cell margin toward the centerand concentrate in the OCS. Fgn-Au particles bindto surface-activated bovine cells, but move very lit-tle, or not at all. All ofthesefactors may contributeto the inability of bovine platelets to react to sur-faces by spreading like human cells, but absence of
the OCS appears to be the major cause. (Am J Pa-thol 1990, 136:399-408)
Human platelet physical transformation after exposureto foreign surfaces in vivo and in vitro has been well char-acterized.1-6 Resting cells are normally discoid in form.After contact with a surface, platelets extend long, finger-like processes and develop a dendritic appearance. Den-dritic platelets undergo conversion to spread forms as thecentral body sinks into the surface and cytoplasm fillsspaces between extended pseudopods.7 Recent studiesin our laboratory using fibrinogen coupled to colloidal gold(fgn-Au) as a probe for glycoprotein llb-lIla (GPIlb-IlIla) re-ceptors have shown that evagination of the surface-con-nected open canalicular system (OCS) is a major factorcontributing to human platelet spreading on surfaces.6
Zucker-Franklin and coworkers9 were first to demon-strate that bovine platelets are similar to human cells inmost respects, but lack the OCS found in human cellsand platelets from most other mammals. As a result, thebovine platelet secretes products of alpha granules byfusion of the organelles to the surface membrane beforeextrusion,10 whereas secretory organelles in human plate-lets fuse with channels of the OCS and use this route fordischarge of products.11
The absence of the OCS in bovine platelets suggestedit might be a useful model for other studies. Because ourrecent investigation identified involvement of the humanplatelet OCS in spreading,6 it seemed reasonable to ex-amine interaction of bovine platelets with foreign surfacesand the binding of fgn-Au receptor probes after surfaceactivation. In contrast to human cells, bovine platelets donot spread, they unfold. After surface activation, bovinecells bind fgn-Au particles but do not move them to cellcenters or the OCS, as human platelets do.
Accepted for publication October 2, 1989.The work submitted is part of the academic thesis (Ph.D.) of Lynette
H. Grouse, B.S.Supported by NIH grants HL-1 1880 and GM-22167 and a grant from
the March of Dimes (1-886).Address reprint requests to James G. White, MD, University of Minne-
sota, Department of Pedeiatrics, Box 198, Mayo Memoral Building, Minne-apolis, MN 55455.
399
400 Grouse et alAJP February 1990, Vol. 136, No. 2
Materials and Methods
General
Blood for the present study was obtained from healthyadult cattle housed under direct supervision of the Schoolof Veterinary Sciences at the University of Minnesota.Samples aspirated from the internal jugular vein were
mixed immediately with citrate-citric acid-dextrose (CCD)(93 mmol/l trisodium citrate, 70 mmol/l citric acid, 140mmol/l dextrose, pH 6.5) in a ratio of nine parts blood toone part anticoagulant.412 Blood was sedimented at roomtemperature under a force of 200g for 20 minutes. Plate-let-rich plasma (C-PRP) above the buffy coat was aspi-rated and transferred to separate tubes.
Human blood was obtained after informed consentfrom young adults who had volunteered for our studieson many previous occasions. None was taking drugs ofany kind at the time or for at least 2 weeks before thestudy. Blood aspirated from an antecubital vein was
mixed immediately with CCD anticoagulant in the same
ratio as for bovine blood, and C-PRP was prepared as
mentioned for bovine samples.
Light Microscopy
Drops of human and bovine C-PRP were placed on poly-lysine-coated glass slides and thin cover slips were
placed on them. The interaction of platelets with polyly-sine-coated glass was viewed under oil through 63X and1 OOX oil-immersion objectives adapted for phase contrastand interference phase contrast observation. Interactionsbetween platelets and glass were followed at intervals for15 to 60 minutes. Between observations the slides weremaintained in a 37 C oven.
Immunofluorescence
Samples of bovine and human C-PRP were mixed with anequal volume of the citrate anticoagulant and centrifugedto pellets. The platelets were resuspended in phosphate-buffered saline (PBS; NaCI, 8 g; KCI 0.2 g; KH2PO4, 0.2g; Na2HPO4, 1.15 g/l adjusted to pH 7.2 with 1.0 mol/INaOH). Twenty microliters of washed platelets were
placed on each of a number of polylysine-coated glassslides, which were inserted in a moist chamber andplaced in an oven maintained at 37 C to settle. At intervalsof 15,30, and 60 minutes, slides were fixed in methanol at-10 C for 10 minutes followed by brief immersions (threetimes for three minutes each) in acetone at -10 C. Astock monoclonal antibody to beta-tubulin (Amersham,
Arlington Heights, IL) was diluted to a concentration of0.01 mg/ml in PBS with 1% bovine serum albumin(BSA).13 Slides were washed once with PBS, and theplatelet spots covered with 20 to 40 ul of the antitubulinantibody. After 30 minutes, the slides were washed withPBS and the spots covered with a fluorescein-conju-gated, anti-immunoglobulin antibody (Cappel, Cochran-ville, PA). After incubation for another 30 minutes, theslides were washed well with PBS and mounted undercoverslips in a solution of p-phenylene- diamine-PBS-glyc-erin using the method of Johnson et al.14 The slides werestudied under phase contrast and phase fluorescence ina Zeiss photomicroscope equipped with an ultravioletpower source and appropriate excitation and barrier fil-ters. Platelets were observed and photographed throughepifluorescence optics and 60X and 1 QOX planapo objec-tives.
Detergent-extracted Cytoskeletons
Samples of detergent-extracted platelets were preparedafter incubation at 37 C for 15, 30, or 60 minutes for ultra-structural study after negative staining according to themethod of Small15 as modified from Hoglund et al.16 TheTriton X-100 glutaraldehyde mixture consisted of 0.5%Triton X-100 and 0.25% glutaraldehyde.13 Detergent ex-traction and fixation were carried out at room temperature.Grids carrying spread platelets were washed briefly inTris-buffered saline, followed by a cytoskeleton buffer(NaCI, 127 mmol/l; KCI, 5 mmol/l; Na2HPO4, 1.1 mmol/l;KH2PO4, 0.4 mmol/l; NaHCO3, 4 mmol/l; glucose, 5.5mmol/l; MgCI2, 2 mmol/l; EGTA, 2 mmol/l; PIPES, 5 mmol/I; pH 6.0-6.1). After washing in the cytoskeleton buffer,the cells were transferred to the Triton X-100 glutaralde-hyde mixture for 1 minute. After a brief wash in cytoskele-ton buffer, the grids were stored on coverslips on thesame buffer containing 2.5% glutaraldehyde for 2 hoursbefore negative staining for electron microscopy. Stainingin sodium silicotungstate was carried out at room temper-ature. Grids were rinsed twice in distilled water and trans-ferred sequentially through four drops of bacitracin (40mg/ml in water; Sigma Chemical Co., St. Louis, MO) in aplastic Petri dish and drained briefly on the edge with filterpaper. They were then passed through four drops of 3%sodium silicotungstate, drained of excess stain, and al-lowed to air dry.
Preparation of Fibrinogen Gold
Kabi grade L and Sigma F4883 fibrinogen preparationswere obtained from commercial sources.17 They were dia-
Bovine Platelets Do Not Spread, They Unfold 401AJPFebruary 1990, Vol. 136, No. 2
lized against deionized distilled water for at least sixchanges and 3 hours. Dialized fibrinogen solutions werethen divided into aliquots, snap frozen, stored at -70 C,and thawed before adsorption. Colloidal gold particleshaving a diameter of 16 to 18 nm were prepared as de-scribed by Loftus and Albrecht.18 One half milliliter of 4%tetrachloroacuric (HAuCI4) acid solution was added to 200ml of deionized distilled water and brought to a boil. Then5 ml of 1% trisodium citrate was rapidly mixed into theboiling solution and the mixture was refluxed for 30 min-utes. The fresh colloid was cooled and pH was adjustedto 6.5 with 0.2 N K2CO3 before use.
The minimum amount of protein necessary to stabilizethe colloidal gold particles was determined by adsorptionisotherms.1920 A series of fibrinogen solutions of increas-ing concentration was made and 0.5 ml of Au (pH 6.5)was added. After 1 minute, 1 ml of 10% NaCI was com-bined and rapidly mixed. The minimum concentration offibrinogen plus 10% of that minimum amount that pre-vented flocculation by NaCI was added to 10 ml of gold.Based on the adsorption isotherm, the average concen-tration of fibrinogen was 6 jg per ml of gold colloid in eachexperiment. After 5 minutes, 0.5 ml of freshly made andprefiltered (0.45 ,m) 1% polyethylene glycol (molecularweight, 20,000) was added to prevent flocculation of thefibrinogen-labeled gold (fgn-Au).1821'22 The fgn-Au wasconcentrated by centrifugation in polycarbonate tubes atan average speed of 1 0,000g for 30 minutes at 4 C. Theconcentrated fgn-Au was separated and resuspended to1 ml with prefiltered (0.2 um) Tyrode's buffer (pH 7.4) withcalcium and magnesium, but without albumin.
Staining with Fibrinogen-Gold (Fng-Au)
Drops of washed platelets were placed on carbon-stabi-lized, formvar-coated grids and allowed to interact withthe surface for 20 minutes at 37 C.17 In some experiments,the incubation time was changed to 15, 30, or 60 minutes.Nonadherent platelets were gently rinsed off the surfacewith HBSS before 10 Al of fgn-Au was applied. After incu-bation with the adherent platelets for 5 minutes at 37 C,unattached fgn-Au particles were removed by rinsing withHBSS. The grids were then fixed in 1% glutaraldehyde incacodylate buffer and air dried for study in the electronmicroscope.
Results
Phase Contrast and Nomarski InterferencePhase Contrast Microscopy of BovinePlatelets During Surface Activation
Examination of bovine platelets after interaction with glassslides for intervals of 15 to 60 minutes by light micro-
Figure 1. Human platelets allowed to interact with a carbon-coated cover slipfor 15 minutes, thenfixed and critical-pointdriedfor study in the electron microscope. Most ofthe plateletshave undergone transformation from resting, discoid cells todendritic shapes with long, spikypseudopods andspreadforms(SP). Magnification X 5000.
scopic methods revealed a significant difference from hu-man cells. Platelets from human subjects may attach to aforeign surface as discoid forms and remain in that statefor only a few seconds up to a minute. The discoid cellsthen extend spikelike pseudopods radiating from a morespherical, central body and become dendritic forms.Slowly the central body sinks into the surface and cyto-plasm fills in spaces between pseudopods, transformingdendritic platelets into spread forms.3.5,6
Bovine platelets during intervals of up to 60 minutesdid not spread on glass surfaces. The cells often re-mained discs for the entire 30-to-60-minute period of ob-servation. Cells that did change shape underwent limitedconversion, often appearing irregular or crescent shaped,but never spread. Some resembled dendritic humanplatelets, but they did not transform into fully spread cellslike their human counterparts.
Scanning Electron Microscopy
Evaluation of human and bovine platelets by scanningelectron microscopy confirmed the observations madeby light microscopic methods.7 23 Human platelets under-went the entire sequence of transformation from discoidplatelets to dendritic forms and then into the thin films typi-cal of spread cells (Figure 1). Bovine platelets also re-sponded to surface activation by changing shape. How-ever, the shape change was limited compared to humancells. Bovine platelets extended long, filiform processes
402 Grouse et alAJP February 1990, Vol. 136, No. 2
Figure 2. Bovineplateletsprepared in the same manner as thecells in Figure 1, but allowed to interact with the coverslipfor 1hour beforefixation. Despite the long exposure, manyplateletshave retained a resting, discoid appearance (RP). Others havetransformed into dendritic platelets (DP), but no spreadformsare apparent. Magnification X 5000.
after contact with the grid surfaces (Figure 2). Their bod-ies flattened to some extent, but never filled the spacesbetween surface extensions. Thus bovine platelets canchange shape and extend filaform processes, but do notdevelop the spikelike pseudopods typical of human den-dritic forms or fill spaces between them to become spreadplatelets.
Immunofluorescence
The unusual response of bovine platelets to glass slideswas reflected by changes in the circumferential microtu-bule. Examination of human cells stained with an antibodyto tubulin and then with a fluorescein-coupled antibody tothe first immunoglobulin revealed uniform fluorescentrings at 15 and 30 minutes of glass exposure (Figure 3).Bright fluorescent rings were also visible at 60 minutes,but in many platelets they developed a disorganized ap-pearance (Figure 4).
Circumferential microtubules in most bovine plateletswere uniform fluorescent circles at 15 and 30 minutes.One hour after exposure to glass most of the rings werestill intact (Figure 5), but some had fractured, leaving be-hind curved, spiral, or straight fluorescent bundles (Figure6). The disorganized patterns observed in human cellswere absent in bovine preparations. On the other hand,the spiral, curved, and straight fluorescent bundles in bo-vine preparations were uncommon in human platelet sam-ples.
Figure&3 Human platelets allowed to interact withbaglass slidefor 15 minutes, then fixed and exposed to anti-tubulin anti-body followed by fluorescein conjugated anti-immunoglobu-lin. Most ofthe microtubule rings appear intact. MagnificationX 630.
Detergent-extracted Cytoskeletons
The failure of bovine platelets to spread was clearly appar-ent in cytoskeletal preparations examined in the electronmicroscope. The thin, pancakelike forms typical of the hu-man platelet response to surface activation were absentin bovine preparations (Figures 7 to 1 1). Discoid cells withcoils of the circumferential microtubule forming the plate-let margin were common (Figures 7 and 8). Crescent- andspindle-shaped platelets with fractured microtubule coilswere also frequent (Figure 9). Other bovine platelets werejagged with broken microtubules or loosened coils ex-tending into surface irregularities (Figures 10 and 11).
Figure 4. Humanplateletsfixed 1 hour after exposure to a glasssurface and the subunit protein of the microtubule, tubulin,stained by indirect immunofluorescence. Some microtubulerings retain a circumferential appearance, while most revealvarious degrees ofdisorganization. Magnification X 630.
Bovine Platelets Do Not Spread, They Unfold 403AJPFebruary 1990, Vol. 136, No. 2
FPba
Figure 5. Bovine platelets exposed to glass for 60 minutes be-forefixation and stainedfor tubulin. The majority ofrings re-main intact 1 hour after glass activation. Other marginal mi-crotubules appear to have fractured. Coils have straightenedinto linear or curved bundles. Magnification X 1000.
Some resembled kites and other irregular forms, but theradiating dendritic shapes and spread forms with a pe-ripheral weave of actin typical of surface-activated humanplatelets (Figure 12) were not present in bovine prepara-tions.
Interaction of Surface-activated BovinePlatelets with Fng-Au Probes
The distribution of fgn-Au probes on surface-activated bo-vine platelets differed strikingly from localization on hu-man cells.17 '8 Neither bovine nor human discoid forms
Figure 6. Bovine platelets fixed 90 minutes after exposure to
glass and stainedfor tubulin. Many rings are intact, but a sig-nificantproportion havefractured and straightened into lin-
ear or curved bundles. There is little difference between thechanges at 60 and90 minutes. Magnification X 1000.
Figure 7. Whole mount of bovine platelets fixed and simulta-neously detergent extracted 30 minutes after exposure to thecarbon-stabilized, formvar-coated grid. Most of the cells haveretained a discoid configuration. Only a few have extendedfiliform processes (FP) in response to surface activation. Mag-nification X 4000.
bound fgn-Au particles. However, dendritic forms in hu-man and bovine preparations were covered with fgn-Auparticles that extended onto pseudopods (Figures 13 to16). As human dendritic platelets transformed into spreadforms, fgn-Au-receptor complexes moved from periph-eral margins toward cell centers and became concen-
trated in channels of the OCS17 (Figure 17).Fgn-Au particles did not appear to undergo reorgani-
zation on maximally altered bovine platelets in the mannerobserved on human platelets. The only apparent move-
ment of ligand receptor complexes was on pseudopods
8
Figure 8. Bovineplateletprepared in the same manner as cellsin the previous illustration 60 minutes after exposure to thegrid. All of the cell membranes have been removed, leavingonly detergent-resistant elements ofthe cytoskeleton and resi-dues of alpha granules. A circumferential microtubule (MT)forms the cell border. The only sign ofactivation is the smallprotrusionfilled with actinfilaments on the left side ofthe cell.Magnification X 17,000.
i. ..
404 Grouse et alAJP February 1990, Vol. 136, No. 2
... .
.! ..!*. :.' ....'i:..
T: F MT
FP
9
Figure 9. Simultaneously fixed and detergent extracted bo-vine platelet prepared 60 minutes after exposure to the grid.The cell has assumed a form resembling a dendritic plateletbut has extended only three filiform processes (FP). The cir-cumferential microtubule coil has fractured and extends inthe long axis of the cell. Microtubule (MT) fragments also ex-tend into the centralprocess. Magnification X 10,500.
and at the edges of some platelet bodies (Figure 18). Fgn-Au particles moved from different locations around thepseudopod to form a linear pattern stretching the lengthof the pseudopod. The linear organization was occasion-ally observed at the inner or outer edge of dendritic bovinecells. However, the central movement of Fgn-Au ob-served routinely on human cells was not seen on bovineplatelets.
Discussion
Bovine platelets, like their human counterparts, are criticalcomponents of the hemostatic mechanism, protecting
I I V'11 ~ ~~~~~~wi.Figure 11. Detergent resistant bovineplatelet cytoskeletonpre-pared 90 minutes after exposure to the grid. The circumferen-tial microtubule (MT) remains coiled in the cell, but elementshave loosened and extend intofiliform processes (FP). The cellhas unfolded to the maximum extent achieved by bovineplate-lets. Magnification X 10,500.
the animal from hemorrhage after vascular injury.2 Yetthere are significant differences between human and bo-vine platelets, and clarifying them may improve our under-standing of platelet function and pathology in humans.Cattle platelets are significantly smaller than human cells,have larger alpha granules, and circulate in larger num-bers.24 Zucker-Franklin and coworkers9 demonstratedthat cow platelets lack the surface-connected OCS pres-ent in similar cells in nearly all other mammalian species.As a result, activated bovine platelets fuse alpha granulesto the surface membrane and discharge products directlyto the exterior10 rather than through channels of the OCS,as in stimulated human cells.1
10
Figure 10. Unfolded bovine platelet simultaneously fixed andextracted90 minutes after exposure to the surface. The micro-tubule (MT) coils have loosened but remain circularfor themostpart. However, elements ofthe marginal microtubule ex-tend into almost every filiform process (FP). MagnificationX 10,500.
= .E'|j_l-.fv.*. . .F"'<
Figure 12. Human plateletfixed and detergent extracted 30minutes after exposure to the grid. The cell isfully spread. Actinfilamentsform a peripheral weave at the margin of the cell.Magnification X 40, 000.
FP I
' ...
Bovine Platelets Do Not Spread, They Unfold 405AJPFebruary 1990, Vol. 136, No. 2
f `#rS-.RP /13-
Figure 13. Bovine platelets allowed to interact with a gridfor20 minutes and then exposed tofibrinogen-gold (Fgn-Au)for5 minutes beforefixation. A restingplatelet (RP) has not boundFgn-Au particles. An adjacent dendritic platelet (DP) has
evenly distributed Fgn-Au over its body andfiliform processes.
Magnification X 10,500.
The present study has shown an even more strikingdifference between human and bovine platelets. Surfaceactivation is a powerful stimulus causing rapid changesin platelet morphology.1`6 Human cells lose their restingdiscoid shape, become relatively spherical, and extendlong, spiky pseudopods as they convert to dendriticforms. Continued interaction causes the spherical body tosink into the surface and cytoplasm to fill spaces betweenpseudopods, transforming dendritic platelets to spreadforms.7 If this process is inhibited significantly in humans,the bleeding time may be prolonged and hemorrhage
25.26may ensue.
Bovine platelets also develop shape changes in re-
sponse to surface activation. They adhere, lose discoid
Figure 14. Bovine platelet exposed to a gridfor 20 minutes andtreated with Fgn-Aufor 5 minutes has manyfiliform processes.Fgn-Au particles are evenly distributed on the many exten-sions. Magnification X 6500.
Figure 15. Two bovine platelets combined with Fgn-Au for 5minutes after exposure to the grid surfacefor 30 minutes. Anearly unfolded cell (1) has a light coating ofFgn-Au particleswhile the other (2) is heavily labeled. Magniflcation X 8000.
form, extend long, filiform processes, and sink into thesurface. However, that is the extent of their reaction. Theydo not develop thicker, spikelike pseudopods or undergoconversion from dendritic to spread forms so typical ofthe human platelet response to surface stimula-tion. 3,5,6,21,22
The variation in cow platelet reactivity27 and spreadingis not known to be associated with any hemorrhagic prob-lem in the animals. Yet the differences in the response ofhuman and bovine cells to surface activation may provideclues to the understanding of human hemostasis. An at-tractive rationale for the failure of bovine platelets to con-
vert from unfolded or dendritic forms to fully spread cellslike their human counterparts is the absence of the OCS.9Recent studies in our laboratory using fibrinogen coupled
S ..l:
*;!! ...' ...,11il:..
.:S
e :.;-' .1! ll.;f..
*,
Figure 16. Unfolded bovine plateletfixed after exposure to thegridfor3o minutes andFgn-Aufor5 minutes. Fgn-Auparticlesare evenly distributed over the body and surface extensions.Magnification X 8000.
-i
k.
. I
DP*:.. .,:... Z:IC
406 Grouse et alAJP February 1990, Vol. 136, No. 2
17Figure 17. Humanplateletplaced on a gridfor2O minutesandthen treated withFgn-Aufor5 minutes. The cell has notspreadfully, but Fgn-Au particles have moved awayfrom the cell mar-gin toward the cell center. They are concentrated in an inter-mediate belt and in channels of the open canalicular system(OCS). Magnification X 8000.
to colloidal gold as a probe for glycoprotein llb-lIla recep-tors have shown that the OCS of human platelets repre-sents a significant membrane reservoir available for ex-pansion of the cell-surface area after activation.8 The ab-sence of the OCS in bovine cells may limit the extent towhich their membranes can expand after surface activa-tion.
The fate of microtubule coils during surface activationof human and bovine cells offers another possible expla-nation for the observed differences.13'26'28 Fluorescencemicroscopy of platelets stained with a monoclonal anti-body to tubulin followed by fluorescein conjugated anti-immunoglobulin demonstrated that circumferential micro-tubules in bovine platelets are more resistant to disassem-bly after long-term surface activation than those in humancells. Nearly all human platelets contained intact microtu-bule coils 15 minutes after exposure to glass, and the ma-jority still possessed them at 30 minutes. However, only afew intact coils remained 60 minutes after human plateletsadhered to the surface.13
All bovine platelets contained coils after 15 minuteson glass and the majority contained coils 30, 60, and 90minutes after activation. Coils that had fractured remainedas bundles in bovine cells. The resistance of bovine plate-let microtubule coils to disassembly on prolonged surfaceactivation may inhibit shape change or limit its extent.When the coil does break it tends to form straight orcurved bundles, which would foster the porpoise andsickle shapes assumed by bovine cells as they respondto glass activation.
Microtubule coils support the discoid shape of restinghuman platelets and contribute significantly to their resis-tance to deformation on aspiration into micropipettes.29
Similar studies have shown that bovine platelets are mark-edly more resistant to deformation in micropipettes thanhuman cells.30 More stable circumferential microtubulesmay be primarily responsible for the resistance, and con-tribute significantly to the inability of bovine platelets tospread on surfaces.
Although microtubule coils remain intact and supportthe discoid shape of most bovine platelets for up to 90minutes after surface activation, the rings loosen and as-sume a ball-of-yarn configuration in cells that develop ashape change. A single microtubule or several polymersare evident in virtually every filiform process extendingfrom the unfolded platelets.10 In contrast, only about 20%of the spiky pseudopods on dendritic human plateletscontain microtubules.12 Yet the filiform processes on un-folded bovine platelets do not develop the appearance ofthicker, spiky pseudopods extending from dendritic hu-man cells. Parallel actin filaments organized in bundlesextend in the long axis of the filiform processes of bovineplatelets, as they do in human platelet pseudopods. Whytheir polymerization and association in cattle platelets failsto produce spiky pseudopods remains obscure.
A similar failure may underlie the absence of spreadingby bovine platelets. Polymerization of actin filaments andtheir association in a peripheral weave at the cell marginis a major factor in the conversion of dendritic humanplatelets into spread forms.31~33The margin of unfoldedbovine platelets is filled with actin filaments, but they donot appear to assemble into the peripheral weave foundin spread human cells.
x|j j.l. .lE .......... ... ..!;.;
.i i;;;!. . .........::v;"V'' :: '
Figure 18. Bovine platelet fixed 20 minutes after incubationon a grid surface and 5 minutes with Fgn-Au. The cell bodyand some filiform processes are lightly stained by Fgn-Au. Onother surface extensions ( ) the Fgn-Au gold particles haveformed a straight line in the long axis. The particles may lie ina groove between closely associatedfiliform processes on somecells, and on single surface extensions on the same or otherplatelets. Magnification X 16,000.
Bovine Platelets Do Not Spread, They Unfold 407AJP February 1990, Vol. 136, No. 2
The differences in cytoskeletal assembly and organi-zation in surface-activated bovine platelets may extend tothe surface membrane cytoskeleton. Fibrinogen coupledto colloidal gold (Fgn-Au) has been used to demonstratethe movement of the glycoprotein llb-lila (GPIlb-lIla) recep-tors on surface-activated human platelets.18'21'22 The elec-tron-dense probes cover the body and pseudopods ofdendritic platelets. As the spiderlike cells convert tospread forms, the Fgn-Au particles move from pseudo-pods and peripheral margins move toward platelet cen-ters and concentrate in channels of the OCS.17
Bovine platelets lack the OCS found in human andmost other mammalian platelets.9 Therefore, the directedmovement of Fgn-Au, GPIlb-llla receptor complexes fromperipheral margins to channels of the OCS on surface-activated human platelets might be expected to differ onbovine cells, and it does. Fgn-Au remained randomly dis-persed on dendritic and unfolded bovine platelets. Theonly significant movement of receptor ligand complexeson bovine cells took place on filiform processes and cellmargins. Fgn-Au particles diffusely covering pseudopodsoften formed linear configurations in the long axis of thesurface extensions and occasionally developed similarlines of gold particles along internal or external margins.The movement of the Fgn-Au probes away from periph-eral margins toward cell centers and the OCS of surface-activated human cells was not observed on the bovineplatelets at any stage of transformation.
The lack of an OCS, however, should not block move-ment of Fgn-Au probes on bovine platelets.' Fully spreadhuman platelets will still transport the ligand-receptor com-plexes toward cell centers, even though channels of theOCS have evaginated back onto the surface8 or haveclosed as a result of the tension caused by spreading.17Preliminary studies in our laboratory have shown that themovement of GPIlb-lila receptor-ligand complexes is notdue to assembly of the internal actin cytoskeleton, butto a cytochalasin B resistant cytoskeleton in the humanplatelet surface membrane.35 Cytochalasin B, however,did not enhance movement of GPIlb-lla receptors on bo-vine platelets.
The present study has shown that bovine platelets donot spread like human cells after surface activation; in-stead they unfold. Absence of an OCS and the reservoirof membrane it represents in human cells for expansionof the surface area during spreading may explain why bo-vine platelets cannot go beyond an unfolded or dendriticphase during surface activation. The inherent resistanceof their circumferential microtubules to disorganization ordisassembly may be, in part, responsible for their inabilityto develop into the spread forms assumed by humanplatelets. While actin filaments polymerize after surfaceactivation of bovine platelets, their organization may be
ineffective in producing spikelike pseudopods and ex-panding the cytoplasm to fill spaces between them. Thevariation in internal cytoskeletal assembly may extend tothe bovine cell surface. Exposure of bovine cells to fgn-Au probes during surface stimulation failed to result in thereorganization observed on similarly treated human cells.This may be due, in part, to absence of the OCS in bovineplatelets. However, it may also be due to inability of GPIlb-lIla receptors to be moved by a cytoskeleton within themembrane of the bovine cell. In that case a difference inthe membrane cytoskeleton may also be a factor in thefailure of the bovine platelet to form an OCS.
References
1. Bizzozero J: Ueber einen neuen Forbestandtheil des Blutesund dessen Rolle bei der Thrombose und der Blutgerinnung.Arch Path Anat Physiol 182,90:261-332
2. Tocantins LM: The mammalian blood platelet in health anddisease. Med 1938; 17:38-155
3. Mattson JC, Zuiches CA: The cytoskeleton of contact acti-vated platelets. Micron 12:69-70,1981
4. White JG: The morphology of platelet function. Meth Hemat1983,8:1-25
5. Loftus JC, Choate J, Albrecht RM: Platelet activation andcytoskeletal reorganization: High voltage electron micro-scopic examination of intact and triton-extracted wholemounts. J Cell Biol 1984, 98:2019-2025
7. White JG: An overview of platelet structural physiology. ScanMicro 1987,1:1677-1700
8. Escolar G, Leistikow E, White JG: Fate of the platelet opencanalicular system (OCS) during surface activation. Circ1988, 78:11546
9. Zucker-Franklin D, Benson KA, Myers KM: Absence of a sur-face-connected canalicular system in bovine platelets.Blood 1985, 65:241-244
10. White JG: The secretory pathway of bovine platelets. Blood1987,69:878-885
11. White JG, Krumwiede M: Further studies of the secretorypathway in thrombin stimulated human platelets. Blood1987,69:1196-1203
12. White JG: Fine structural alterations induced in platelets byadenosine diphosphate. Blood 1968, 31:604-622
13. White JG, Sauk JJ: Microtubule coils in spread blood plate-lets. Blood 1984, 64:470-478
14. Johnson DG, de C Nogueira-Araujo GM: A simple methodof reducing the fading of immunofluorescence during mi-croscopy. J Immunol Metho 1981, 43:349-355
15. Small JV: Organization of actin in the leading edge of cul-tured cells: Influence of osmium tetroxide and dehydrationon the ultrastructure of actin meshworks. J Cell Biol 1981,91:695-704
408 Grouse et alAJP February 1990, Vol. 136, No. 2
16. Hoglund AS, Karlson R, Arro E, Frederiksson BE, LundbergU: Visualization of the peripheral weave of microfilaments inglial cells. J Muscle Res Cell Motil 1980,1:127-146
17. Leistikow EA, Barnhart Ml, Albrecht RM, White JG: Redistri-bution of fibrinogen receptors on surface activated platelets.In: Fibrinogen 3. Biochemistry, Biological Functions, GeneRegulation and Expression. Edited by Mossesson MW. NewYork, Elsevier Publishers, 1988, pp. 215-220
18. Loftus JC, Albrecht RM: Redistribution of the fibrinogen re-ceptor of human platelets after surface activation. J Cell Biol1984,99:822-829
19. Geoghegan W, Ackerman GA: Absorption of horseradishperoxidase, ovomucoid and anti-immunoglobulin to colloidalgold for the indirect detection of ConA, wheat germ aggluti-nin and goat anti-human immunoglobulin G and cell surfacesat the electron microscopic level: A new method, theory andapplication. J Histochem Cytochem 1977,25:1187-1200
20. Horisberger MG, Rosset J, Bauer H: Colloidal gold granulesas markers for cell surface receptors in the scanning elec-tron microscope. Experientia 1975, 31:1147-1149
21. Goodman SL, Albrecht RM: Correlative light and electron mi-croscopy of platelet adhesion and fibrinogen receptor ex-pression using colloidal-gold labeling. Scan Micro 1987, 1:727-734
22. Oliver JA, Albrecht RM: Colloidal gold labeling of fibrinogenreceptors in epinephrine and ADP-activated platelet suspen-sions. Scan Micro 1987,1:745-756
23. White JG, Krumwiede M: Influence of cytochalasin B on theshape change induced in platelets by cold. Blood 1973, 41:823-832
24. Gentry PA, Downie HG: Blood coagulation. In: Dukes' Physi-ology of Domestic Animals. Edited by Swenson MJ. NewYork, Cornell University Press, 1984, pp. 41-50
25. White JG, Gerrard JM: Ultrastructural features of abnormalblood platelets. Am J Pathol 1976, 83:590-632
26. White JG: Platelet microtubules and microfilaments: Effectsof cytochalasin B on structure and function. In: Platelet Ag-gregation. Edited by Caen J. Paris, Masson and Cie, 1971,pp. 15-52
27. Dodds WJ: Platelet function in animals: species specificities.In: Platelets: A Multidisciplinary Approach. Edited by de Gae-tano G, Garattini S. New York, Raven Press, 1978, pp. 45-59
28. White JG, Rao GHR: Influence of a microtubule stabilizingagent on platelet structural physiology. Am J Pathol 1983,112:207-217
29. White JG, Burris SM, Tukey D, Smith C, Clawson CC: Mi-cropipette aspiration of human platelets: Influence of micro-tubules and actin filaments in deformability. Blood 1984, 64:210-214
30. Smith CM II, Burris SM, Weiss DJ, White JG: Comparisonof bovine platelet deformability using micropipette elasticity.Am J Vet Res 1989, 50:34-38
31. Karlsson R, Lassing I, Hoglund AS, Lindberg U: The organi-zation of microfilaments in spreading platelets: A compari-son with fibroblasts and glial cells. J Cell Physiol 1984, 121:96-113
32. White JG: Arrangements of actin filaments in the cytoskele-ton of human platelets. Am J Pathol 1984, 117:207-217
33. Escolar G, Krumwiede M, White JG: Organization of the ac-tin cytoskeleton of resting and activated platelets in suspen-sion. Am J Pathol 1986,123:86-94
34. Behnke 0, Bray D: Surface movements during the spreadingof blood platelets. Eur J Cell Biol 1988, 46:207-216
35. White JG, Escolar G, Leistikow EA: GPIlb-Illa receptor mobil-ity within membranes of surface and suspension activatedplatelets does not depend on assembly and contraction ofcytoplasmic actin. Clin Res 1989, 37:603A
