(CANCERRESEARCH55, 792—798,
February15,1995]
Behavior of Crocidolite Asbestos during Mitosis in Living Vertebrate Lung
Epithelial Cells1
Jeffrey G. AUIt,2 Richard
W. Cole, Cynthia G. Jensen, Lawrence
C. W. Jensen, Lori A. Bachert,
and Cozily L Rieder
Laboratory of Cell Regulation, Division of Molecular Medicine, The Wadsworth Center, New York State Department of Health, Albany, New York 12201-0509 1). G. A.,
R. W. C., L A. B., C. L RI; University ofAuckland@ Private Bag 92019, Aucklan4 New Zealand (C. G. J., L C. W. ii; and Department ofBiontedicoi Sciences, State University
ofNew York, Albany, New York 12222 (C. L R.J
ABSTRACT
‘Asbestos
has beendescribedasa physicalcarcinogenIn that long thin
fibers are generally more carcinogenicthan shorter thicker ones.It has
been hypothesized that long thin fibers disrupt chromosome behavior
during mitosis, causing chromosome abnormalities which lead to cell
transformation and neoplastic progression. Using high-resolution time
lapse video-enhanced liejd microscopy and the uniquely salted lung epi
thelial cells of the newt TariCIta granulosa, we have Characterized for the
sion occur in asbestos-induced
cancers,possiblyby activating onco
first time the behavior ofcrocidollte asbestosfibers, and their interactions
with chromosomes, during
mltosis In living cells. We found that the
keratin cage surrounding the mitotic spindle Inhibited fiber migration,
resulting In spindles with few fibers. As in interphase, fibers displayed
microtubule-mediated
saltatory movements. Fiber position was only
slightly affected by the ejection forces of the spindle asters Physical
interactions between croddolite fibers and chromosomes occurred ran
domly within the spindle and along its edge. Croddolite fibers showed no
genesandeliminatingtumorsuppressorgeneswithin the genome(14,
18). It hasbeenhypothesizedthat asbestosmay causechromosomal
abnormalities by interfering with chromosome segregation during
mitosis (19, 20). The incidence of chromosomelaggards, chromo
some bridges, and sticky chromosomesduring anaphaseincreases
with asbestostreatment (21—23).Observations on fixed material
have shown asbestosfibers near and, in some cases,penetrating
affinity toward chromatic and most encounters ended with the fiber
chromosomeswithin the spindle (17, 21, 24).
Sincethe ability of a fiber to transforma cell is relatedprimarily to
its size and shape, and not its chemistry, asbestosis considered to be
passively yielding to the chromosome. In a few encounters along the
spindle edge the chromosome yielded to the fiber, which remained sta
tionary as if anchored to the keratin cage. We suggest that fibers thin
enounja to be caught In the keratin cage and long enouaji to protrude
of the surface of a fiber are all believed to contribute to disease
development(8).
Asbestos-induced mesotheliomas and bronchogenic carcinomas are
characterized by specific types of nonrandom chromosomal abnor
malities, which include deletions, translocations, and aneuploidy of
specificchromosomes(ii, i3—i7).These abnormalitiesare believed
to be the mechanismby which cell transformationand tumorprogres
a physicalcarcinogen(25). Long (>4 p.m),thin (<0.25 pm) fibers arc
into
the spindleare thosefibers with the ability to snagor block moving generally more tumorigenic than shorter, thicker ones (Ref. 26; re
viewed in Refs. 1, 27). Reducing the length of a fiber by milling
chromosomes.
reducesits carcinogenicity(28, 29). The importanceof fiber dimen
sionsto carcinogenicityis not simply related to fiber aerodynamics
INTRODUCTION
andthe selectiveexpulsionof certain-sizedfibers from the lung or to
theability of certain-sizedfibersto penetratethelung lining, sincethe
Epidemiological studies and experiments on laboratory animals
have long linked asbestosexposureto various diseases.Crocidolite
fibers in particular have been associated with the development of
pulmonary interstitial fibrosis (asbestosis),lung cancer, and malignant
mesotheliomasof the pleura,pericardium,and peritoneum(reviewed
cultured cells or implanted into the pleurae of rats (ii,
asbestos exposure and cigarette smoke have demonstrated
fibers accumulatein the perinuclearregion, placing them at the time
same relationship
is also seen when various-sized
fibers are added to
i9, 29). At the
cellular level, comparedwith shorterfibers, longer fibers are more
easily phagocytized(19), but less easily transportedalong microtu
bules
once inside the cell (30, 31). However, both long and short
in Refs. 1—3).
In the caseof lung cancer,but not mesotheliomas,
a multipli
cative effect (4). The mechanismsby which asbestosfibers cause
disease remain largely unknown (5—7).There is strong evidence that
asbestosis a complete carcinogen,both initiating and promoting
tumorigenesis(reviewed in Ref. 8). However, unlike most carcino
gens,asbestosby itself doesnot appearto be mutagenicas is evident
in a variety of gene mutation assays(9—12).
Alternatively, it does
induce both structural and numerical chromosomal abnormalities in a
wide range of mammalian cells, including mesothelial cells (ii,
13—17).The ability of asbestos to cause a variety of different diseases,
affect different target cells, induce cell transformation, and promote
tumor growth indicates that asbestosmust have multiple mechanisms
of action.The dimensions,durability,andthe physicochemicalnature
of mitosis in a position to interactwith chromosomeson the spindle
(5, 19). It has been hypothesized
that long thin fibers may affect
chromosomesegregationsterically, either by directly blocking chro
mosome movement to a pole or by becoming entangledin the spindle
apparatus(21,2Z 24, 32).Long fibersfoundwithin thedeavagefurrow aftez
the completionof karyokinesisindeedhavebeenshown to affect chromo
some distribution sterically by inhibiting the completion of cytokinesis,
resultingin tetraploidor binucleatedcells (i7, 22, 31, 33).
A numberof otherpossibleroutesfor asbestos-induced
aneuploidy
havealsobeenproposed(reviewedin Ref. 20). For example,certain
typesof asbestosfibers may havean affinity, throughcharge-charge
interactions,for specific componentsof the chromosomeor spindle
apparatus(ii, 21, 33). In this case,fibers would disruptchromosome
Received10/6/94;accepted12115/94.
Thecostsof publication
of thisarticleweredefrayedin partby thepaymentof page
charges.This articlemustthereforebe herebymarkedadvertisementin accordancewith
18 U.S.C.Section1734solely to indicatethis fact.
segregation by sticking to chromosomes or to the spindle apparatus,or
by causing defective chromosomesor spindles by absorbing the
componentsbefore they assemble.Alternatively, long thin asbestos
fibers may be more likely to disruptthe phagolysosomalmembrane
by NIH GMS Grant R01-40198(C. L R.), andby National Centerfor ResearchResources that surroundsthem, releasinghydrolyticenzymesinto the cytoplasm
GrantRR01219(Department
of HealthandHumanServices/Public
HealthService)to
cells
support the Wadsworth Center's Biological Microscopy and Image Reconstruction (34) and possiblypoisoningthe cell. Chromosomesof “sick―
Facility as a National Biotechnological Resource.
indeed tend to stick together and form bridges during anaphase.3
, This
work
2 To whom
was
supported
requests
by American
for reprints
Lung
should
Association
be addressed,
Grant
RG-018-N
at the Laboratory
(J. G. A.),
of Cell
Regulation,Divisionof MolecularMedicine,The WadsworthCenter,Albany, NY
12201-0509.
3 Unpublished
observations.
792
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
1@@@.
@
-
CROCIDOLITEASBESTOSAND MITOSIS
To determine how various-sizedasbestosfibers interact with the
chromosomesand the spindleduring mitosisit is necessaryto follow
their behavior within living cells amenableto high-resolutionoptical
analyses. In this context we have shown previously that the lung
epithelial cells of the newt, Taricha granulosa, are ideally suited for
studyingthe in vivobehaviorof asbestosfibersduringinterphase(30).
The large size, flatness,and optical clarity of thesecells allow one to
visually track individual asbestosfibers and to observe how they
interactwith variouscell components.Importantly, newt lung epithe
hal cells have the same basic fundamental structural organization,
biochemical pathways, and mechanisms supporting cellular processes
as their counterparts in humans. In this report we detail for the first
time how crocidolite asbestosfibers behave within living cells during
mitosis.
MATERIALS
TQ-2028 optical memory disk recorder (ADCO Aerospace, Fort Lauderdale,
FL). Cellswereviewedat 23°C
usingheat-filtered546 nm light. During time
lapse microscopy, images of cells were captured and recorded every
s, and
the illumination was shuttered between each exposure. For analyses of fiber
lengthsand ratesof motion, sequentialimagesfrom theserecordingswere
played back, image by image,through a time-basecorrector(For.A Corp.,
Natick, MA) into Image-i, which contains the appropriate analytical programs.
For producingfigures,selectedvideo imageswerephotographedonto Plus-X
film (EastmanKodakCo.,Rochester,NY) usinga freeze-framevideorecorder
(Polaroid Corp., Cambridge,England). This film was then developedin
Rodinal (AGFA Corp., Ridgefield Park, NJ).
Immunofluorescence Light Microscopy. The procedures for the indirect
immunofluorescent
staining of microtubules
(36, 37) and keratin (38) shown in
Fig. 1 have been describedpreviously. The monoclonal antibody against
@3-tubulin(TU-27B) and the polyclonal antibody against keratin (Ku) used
were kind gifts from Drs. L. Binder (University of Alabama, Birmingham, AL)
and B. S. Eckert (State University of New York, Buffalo, NY), respectively.
AND METHODS
CellCultures.Primary
cultures
of T.granulosa
lungepithelial
cellswere RESULTS
derivedfrom small lung fragmentsasdescribedpreviously(35). Briefly, lung
fragments were cultured at 23°Con 25-mm2 glass coverslips in Rose multi
purpose chambers filled with half-strength L-15 medium supplemented with
Intermediate
10%fetalcalf serumandantibiotics(36).Within 7—10
days,monolayersheets
of epithelial cells migrateup to 1—2
mm from the tissuefragments.At this
time, the medium was replaced with fresh medium containing 0.5 mg/mI of
InternationalUnion AgainstCancercrocidoliteasbestos(a kind gift from Dr.
T. M. Fasy, Mount Sinai Medical Center, New York, NY). As reported
previously (30) numerous crocidolite fibers become incorporated by phagocy
tosis into 1'.granulosaepithelialcells within the first 24 h of exposure.For
high-resolutionmicroscopy,coverslipswith mitotic cells containingasbestos
fiberswereremovedfrom the Rosechambersandmountedon microperfusion
chambersfilled with freshmedium(seeRef. 30).
Video-enhanced Light Microscopy. The video-enhancedlight micro
scopicsystemusedconsistedprimarily of a Nikon Microphot-FXdifferential
interference contrast light microscope equipped with X60 (numerical aper
tore = 1.4) and X40 (numerical aperture = 0.85) objectives. This microscope
was coupled to a Hamamatsu C2400 Chalnicon camera, the output of which
was routed through an Image-i image processor(Universal Imaging Corp.,
West Chester,PA). Image processingwas used to eliminate background
optical noise within the system and to enhance the signal:noise ratio by
averagingtheimageover 16frames.After processing,theimagewasportedto
and stored on Panasonic TQ-FH224 optical memory disks using a Panasonic
p. H
@
@
@
@
@
@
@
@
Fig. 1. Differential interference-contrast
(A, D),
antikeratinimmunofluorescent(B, if), and antitu
(F). Bar, 30 p.m.
Most Fibers
.
from
Interacting
.@.:‘-
-
:‘@@X
,@-‘
\).
bulin immunofluorescent(C, F) micrographs of
newtlung epithelialcells.The majority of incorpo
rated crocidolite fibers (arrows) accumulatenear
the nucleusduring interphase(A) but remainout
side the spindleregionduring mitosis(D). During
the transitionfrom interphaseto mitosis,a cageof
keratinintermediate
filamentssurrounding
the in
terphasenucleus continuesaround the prophase
nucleus(B) andaroundthe mitotic spindle(if). In
contrast,the long cytoplasmicmicrotubulesof in
terphase(C) disassembleand are replacedby the
shorterastral microtubulesof the mitotic spindle
Prohibit
‘@-,.@- .
.
@
Filaments
with the Forming Spindle and Chromosomes. After incorporation
into an interphasecell, the majority of asbestosfibers are transported
to and accumulate around the nucleus (Fig. 1A; Refs. 5, 19, 30), where
they remain throughoutthe prophasestage of mitosis. Despite this
fact, we found that the forming spindle remained relatively free of
fibers (Figs. 1D and 2—5).
The spindle in vertebrate cells forms in that
region previously occupied by the nucleus. In epithelial cells, this
region is often substantially larger than the spindle, and it is optically
clear in that it is virtually devoid of mitochondria and other large
organelles (Fig. 2; Ref. 38). It has been shown previously that this
clear area is created by keratin intermediate filaments, which form a
“cage―
around the interphase and prophase nucleus (Fig. 1B). In many
cell types this cage persistsfor various periodsof time after nuclear
envelope breakdown (Fig. 1E; Ref. 38) and prohibits the migration of
large organelles and cell inclusions, including asbestosfibers, into the
region of the forming spindle. As mitosis progresses,the keratin cage
usually slowly collapses around the spindle (38) so that by the time
spindle formation is complete (i.e., metaphase), mitochondria and
other excluded bodies may come to lie very close to the spindle
. , )
,“•4
:;
.
.
-
‘I;c..
,@i+
-
A
______________________________
,,@):,;
L@'
,,@
•
‘
@(‘
/V
f:
,J
•I@@'
“
j
I@/‘
,@.@Ii@EF@F@@-,
,
793
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
.
.,..
0'-
@
‘.
,
CROCIDOLITEASBESTOSAND MITOSIS
@
.@ 7 - ‘ ‘1';@)@
.-@,.:@,
@
@
.----.
@
,
@:-
,,@.—.——::.:@@
,..-.
!
1,,
“
@/
@
@
1,1
@
@
@
@
@
@
@
@
@
@
@.1
.@
V
@/@/
@fd
(
\
J,.;@J•#@@;;
tb.;
;/
‘;@
..-@‘-:
I
I
I
,@•z':@@‘ @‘@; .:@,.;
.1;fA@
‘111
@
@
•@‘(
:‘
@•,.•
..@“
,,@.,.@@rA;4
.4
a,,-‘
/.
1@' ‘
I
‘1
4
@.
,
‘@
‘@ •
, •f ,.
p
-- @v./
@: •@‘;i@i.
@‘
;“@@;,;@@1 @v4i@
‘@@)
#@;
.
• ;
:@ ....‘
, ,@‘@•;
1@@/'
- -
,
@.
,.?
@1 (I
:;@,
•
,@
@.I,J.-,
..@‘IJ
‘.@
r__@.
-‘
-i_f
I
,
@. -..@,
@
@- ‘ F@
:@r
@L'@
‘@
‘@i1i;X:::)
@‘
f.
1@
,:d.—@::
,@
-I
,;!
,@
;@ :@“@:@
;.@
@—J;.,
a
,2'
l•
j,,.
,@
F
,.
.-
,tj@
..,.,
,
-.
‘ .,
.
@
@
•
,@.ø I..,
•“
•1
..)
@
1-
:@. “@@‘
@
@
,@.
,
J
•
,.@
1_4
-•‘A
Fig. 2. Selectedmicrographs,from a time lapsevideo-enhanced
differentialinterferencecontrastrecordingof a formingmitotic spindle.Thespindleis locatedin a cleararea(small
arrowheads)producedby a keratincage(0 mm). As mitosisprogresses,the keratincagecollapsesaroundthespindle,allowing mitochondriaandcrocidolitefibers(large arrow) to
entertheclearareaandto movecloserto thechromosomeson the spindle(47—78
min). A crocidolitefiber (large arrowhead),12.9p.mlong and0.4 p.mwide, displayingsaltatory
motionsexhibiteduniquebehaviorin that it penetratedthrougha chromosome(small arrow) anddraggedit poleward.The time lapsesequenceshowsthe transportedfiber (large
arrowhead) entering the spindle region (0 rain), moving out of the spindle (34 mm), moving toward the spindle pole with the snagged chromosome (37, 47, and 51 min), turning away
from the pole (56 mm), the freedchromosome(smallarrow) expelledfrom thespindleby the astralejectionforce(63 min), andthe fiber without theattachedchromosomemoving
close to the pole (78 mm). Penetrationof the fiber throughthe chromosomeis particularlyevident at 47 and 56 mm. Time (in mm) in upper right corner. Bar, 15 p.m.
periphery. During this time, asbestos fibers within the keratin cage
also moved with the cage to the edge of the spindle and closer to the
chromosomes (Fig. 2). However, since the majority of these fibers
never completely pass through and escapethe keratin cage, relatively
few fibers were actually seenassociatedwith the spindleor its polar
regions.
Crocidolite Fibers Exhibit Microtubule-mediatedTransport
during Mitosis. In vertebrates, the mitotic spindle is assembledafter
nuclear envelope breakdown from an interaction between the chro
mosomes and spindle poles, the latter of which consist of two radial
microtubule arrays known as asters (Fig. iF). For this study we have
defined the spindle to be composed of its associated asters and the
microtubule-rich
region between the asters containing
the chromo
somes and spindle proper. Once within the spindle, crocidolite fibers
sometimes displayed saltatory movements similar to those previously
thoughtto be causedby the continuousgrowth of microtubuleplus
ends away from the aster (36). We found that the position of crocid
olite fibers was only slightly affected by the ejection properties of the
asters.Over time some of the fibers that were not actively tethered to
and moving along astral microtubulestendedto migrate away from
the astral centers(Fig. 3; 50—88mm). This expulsionof fibers was
not as prominent or as efficient as that of chromosomefragments,
which are ejected in a slow (—0.04sm/s) but relatively continuous
motion (40). When it occurred,the ejectionof crocidolitefibers took
longer and fiber motion was discontinuous.Fibers which pointed
directly toward the astralcenter,i.e., thosethat were arrayedparallel
to astral microtubules, were less likely to be influenced by the astral
ejectionforce (Fig. 4).
Physical Encounters between Crocidolite Fibers and Chromo
somes Usually
Result
in Fiber
Displacement.
Crocidolite
fibers
within the spindleand along its edgewere thosefibers in positionto
was apparent that the microtubule-mediated transport of membrane
physicallyinteractwith chromosomeson the spindle.Although rela
bound fibers observedin interphasecells (30) also occurredduring tively rare, chromosomefiber encountersdid occur(Figs. 2—5).
Such
mitosis. In mitotic cells these movements were directed along astral encounterswere by chance and occurred either when a moving
microtubules radiating from the centers of the two spindle poles. The chromosomebumpedinto a stationaryfiber or when a moving fiber
averagerate of transport [0.46 ,tm/s; 24 particles undergoing a total of contacted a stationary chromosomearm. During these encounters
crocidolite fibers never stuck to, nor showed any affinity for, the
35 ±0.22 (SD) translocations] was not significantly different from
that seenduring interphase [0.48 sm/s; 25 particles undergoing a total chromosome (Fig. 3). Most encounters ended with the fiber yielding
of 32 ±0.25 (SD) translocations]data from (Ref. 30) (P = 0.86; t to the chromosome(Table 1); i.e., either the moving chromosome
distribution, df = 65). The maximum velocity observed was 0.98 ,.&m/s pushedthe fiber aside(Figs. 3 and4) or the moving fiber bouncedoff
compared with 1.18 p.m/s during interphase (30). The average length the chromosome.In the 10 casesin which a stationaryfiber did not
of the transporting fibers studied was 4.5 @.tm
with a range of 1.3—12.9 yield to the chromosome, the fiber was positioned along the edge of
the spindle presumablyanchoredin the keratin cage. In 9 of these
@m.
cases, the chromosome squeezed around the fiber; in the remaining
Ejection Properties of the Asters Only Minimally Affect Fiber
case, the chromosome was momentarily snagged on the fiber and its
Position. An astralejectionforce existsin mitotic vertebrateepithe
movement was temporarily delayed.
hal cells that expels chromosome fragments and other large bodies
We observeda chromosomeyielding to a moving fiber only twice
away from the centers of the two polar asters(39). It is dependent on
the dynamically unstable behavior of astral microtubules and is (Table 1). In one of these cases, a chromosome arm was displaced
794
reported to occur along microtubules
in interphase cells (30, 31). It
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
@
@‘
,
CROCIDOLITEASBESTOSAND MITOSIS
@
@
“a;.,
\@ç'
.t―i
‘
t1@4
@
‘p
@‘‘.4k
,
.
@c,
@
‘
‘
@@A''
@‘@)@‘
‘y
Fig. 3. A crocidolite fiber within the mitotic spindledemonstratingdifferent typesof fiber movement.The fiber, 12.8p.mlong and0.4 p.mwide, beganthe sequencepositioned
betweenthechromosomesat theequatorialplateanda spindlepole,with its lengthrunningparallelto thespindleaxis(0 min). A lysosomalbulge(arrowhead)alongthefiber indicated
that the fiber is membranebound. Fiber movementcommencedduring the first 37 mm when a chromosomearm (large arrow) pushedthe fiber aside (10 mm). From 37 to 50 mm,
thefiber wasactivelytransportedtowardthespindlepole(smallarrow). The fiber wasthenpushedawayfrom thepole(88 mm) by the astralejectionforce.During anaphase(88-99
min), thefiber wasagainpassivelypushedasideby thesegregatingchromosomesof theelongatingspindle.Althoughcloseto thechromosomes,
thefiber wasexcludedfrom theforming
nucleus(117 mm). Time (in mm) in upper right corner. Bar, 15 p.m.
slightly by the fiber. In the other case, the moving fiber penetrated
through the chromosome and dragged the whole chromosome, which
had not yet attached to the spindle, toward the center of the aster (Fig.
birefringence from the many fibers prevented the identification of
individual fibers within this region. However, as one chromosome
2). As the fiber and chromosome
suddenly appeared (Fig. 5; 37 mm). During anaphase, the resultant
moved poleward, they met resistance
from the astral ejection force (Fig. 2; 51 mm). The fiber eventually
turned away from the pole (Fig. 2; 56 mm), and the chromosome
slippedoff the fiber. The chromosomewas thenexpelledby the astral
ejection force to the periphery of the spindle (Fig. 2; 63 mm). The
fiber, now lacking the attached chromosome, moved closer to the
center of the aster (Fig. 2; 78 mm). The chromosomeultimately
attached to the spindle and moved poleward on its own without the aid
of the fiber.
CrocidoliteFiber-ChromosomeContactsMay InduceChromo
some Breaks.
In one cell a crocidolite
fiber
appeared to break a
chromosome(Fig. 5). In this cell, rotation of the spindle dragged
chromosome arms along the edge of the keratin cage and into a region
containing
@
@
numerous fibers (Fig. 5; 37 mm). Unfortunately,
extensive
•.;///Pj/;:@
,.il
f(
arm was dragged through
acentric
chromosome
this region,
fragment,
a break in both chromatids
consisting
of both chromatids,
was
left behind at the equatorial plate of the spindle (Fig. 5; 58 mm). The
fragment ended up in the cytoplasm of one of the daughter cells
(Fig. 5; 105 mm).
DISCUSSION
This study is the first to describe the behavior of phagocytized
crocidolite asbestosfibers during mitosis in a living cell. Considering
the long latency period between asbestos exposure and disease in
humans, we were not surprised that most crocidolite-containing cells
completed mitosis normally. During this study, cells which had mod
erate (—10—20)
numbers of fibers showed no signs of cytotoxicity,
o•
;@‘::.‘
@;
,,@
j.',
Fig. 4. A largecrocidolitefiber at a spindlepole during mitosis.A membrane-bound
fiber (arrowheadpointsto the lysosomalbulge), 16.0p.mlong and0.6 p.mwide, beganthe
sequencepositionedat oneof the spindlepoles(0 mm). The fiber remainedwith the spindleasthe spindlerotated(0—48
mm). During the rotation,thespindlepole movedfrom the
midpoint of the fiber (0 mm) to the endof the fiber oppositethe lysosomalbulge(27 mm). This relativepositionof the fiber to the pole continued(48 mm) until anaphase,during
which time the fiber was passivelypushedasideby the elongatingspindle(106 mm). Time (in mm) in upper right corner. Bar, 15 p.m.
795
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
@
@;
::
i@
i
@i
CROCIDOLITE
ASBESTOS
AND MITOSIS
However, the microtubule-mediatedmotion of fibers during inter
Table 1 Chromosome-fiberencounters(146
cells)Fiber
encountersin 29
touch)ChromosomeTotal
Chromosome
yields
yields
115
phase appears to be size dependent in that fibers >5 p@mrarely
exhibited saltatory motions (30). By contrastwe found that longer
fibers (up to 12.9 p.m) were frequently transported during mitosis
(Fig. 3). This difference in fiber behavior during interphase and
mitosis can be explained by the following assumptions. Phagoly
sosomes containing long fibers are in contact with many more
microtubules than phagolysosomes containing short fibers because
of their greater surface area. As a result their molecular motors
interact with many more microtubules at any given time. During
Neitheryields
(slight
95 (82.6)― 10k'(8.7)
10 (8.7)
intofiberFiber
moves
moves
31
intochromosomeTotal
10a
29 (93.5)
146
2c (6.5)
124
0(0)
12
Numbersinparentheses,
percentage.
b Chromosomes
squeezed
around
fiber.
In one
case,
the
chromosome
was
interphase,
snagged
momentarilyon the fiber.
C In
one
case,
the
chromosome
was
slightly
displaced;
in
the
other
case,
microtubule
density
is high but neighboring
microtu
bules are often oriented in many different directions (Fig. 1C). The
molecular motors on the surfaceof a long fiber can be expected to
interact with and move along many microtubules with different
orientations. Since the fiber is pulled in many directions simulta
neously, it will exhibit little net motion. By contrast, each of the
two half-spindles in a mitotic cell contain highly polarized arrays
of near parallel microtubules (Fig. iF). Under this condition the
molecular motors on the surface of a long fiber can be expected to
interact and move along many microtubules oriented in the same
the
chromosomewas snaggedand dragged by the fiber.
such as excessive vacuolization, collapsed spindles, or sticky chro
mosomes. This may be due in part to the phagolysosomal membrane,
which effectively isolates each crocidolite fiber from the rest of the
cell, and which probably prevents physicochemical reactions between
the fiber surface and potential intracellular targets including chromo
somes. Contrary to earlier reports (41, 42; reviewed in Ref. 43), we
generaldirection, resulting in movementof the fiber in that direc
have recentlyfoundthat mostif not all fibers remainsurroundedby a
tion. Under this scenario, saltatory movement of longer fibers
membrane weeks after their incorporation into vertebrate epithelial
could occur during interphase in cell types with less complex
cells (31; seealso Ref. 44). Indeed, many of the fibers we observed in
microtubule networks (31).
mitotic cells by differential interference contrast microscopy dis
We found that the astralejectionforce, which expelschromosome
played an obvious membrane bulge, which indicated that they were
fragments and large bodies away from the two spindle poles, had only
still enclosed within an intact phagolysosome (Figs. 3 and 4).
a minimal effect on crocidolite fibers. On the basis of what we know
Phagocytized crocidolite fibers undergo saltatory motions that ul
regardinghow this ejection force works, this was to be expected.
timately lead to their perinuclear accumulation (30, 31). This “salta
Current evidence suggeststhat the mechanism for producing this force
tory―type of motion is characteristically exhibited by other non-fiber
containing phagosomes and phagolysosomes as they are transported is based on the constant outward growth of dynamically unstable
of.
along microtubules within the interphase cell (45—47).This motion is microtubule plus ends, which in effect creates a steric “wall―
microtubule ends that are constantlymoving away from the astral
mediated by several different classes of membrane-bound mechano
chemical (molecular motor) proteins which move organdIes along center (36). Other data suggestthat the magnitude of the ejection force
microtubulesurfacesin different directions,dependingon the type of dependsprimarily on the cross-sectionalsurfaceareaof theobjectthat
motor (48—50).As the cell enters mitosis the relatively stable, long, facesthe polar microtubulearray; i.e., the smallerthe cross-sectional
curvy cytoplasmic microtubules of interphase (Fig. 1C) disassemble
and are replacedby more dynamic,shorter,straighterastralmicrotu
bules that ultimately form the spindle (Fig. iF; see Ref. 51). In this
study, we found that the saltatory motions exhibited by crocidolite
fibersduringinterphase(30) continueduringmitosis.We attributethis
behaviorto the abundanceof microtubuleswithin the mitotic cell and
to the presence of membrane-based microtubule motors around the
fiber. Indeed, the average rate that fibers moved along astral micro
tubules was similar to that along interphase cytoplasmic microtubules
(0.46 p.m/s during mitosis versus 0.48 nm/s during interphase).
surface area the lesser is the magnitude
of the ejection
force (40).
Becauseastralmicrotubulesare radially arrayed,the further an object
is from the astral center, the lower is the density of microtubules that
it encounters.As a fiber at the peripheryof the astralarray interacts
with and moves along a microtubule toward the astral center (via
molecular motors), resistanceand drag will align it parallel to the
microtubule, at which point it presentsits minimal cross-sectional
surface area to the pole. Therefore, thin fibers would present a
relatively small target for the outward growing microtubulesof the
aster.Of course,the expulsionof crocidolitefibers from the asterby
rotated,chromosomearmswere draggedthrougha regionwith numerousfibers. A chromosomebreak (arrowhead) was observedin one of thesearms shortly after this rotationwas
completed(37 mm) andtheresultingchromosomefragment(arrow) wasleft behindat anaphase(58 min). The fragment(arrow) wasexcludedfrom eachof thedaughternuclei(105
mm). Time (in mm) in upperright corner.Bar, 15 p.m.
796
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
CR0CIDOL@FEASBESTOSAND MITOSIS
growing
microtubules
may also be inhibited
by attachments
(via
molecular motors) of the fiber to neighboringmicrotubules.
It hasbeen hypothesizedthat crocidoliteasbestosmay affect chro
mosome segregationand that this is a stochasticprocessbased on
stericconsiderations(21, 22, 24, 32). If true, the inability of our study
to clearly demonstratethe disruptionof chromosomesegregationby
crocidolite can be attributed to the fact that only very few of the many
fibers within the cell are actually able to pass through the keratin
intermediate filament cage to contact the spindle and that when they
do they are not normally in a positionto interfere substantiallywith
bly due to the breakdownof the keratin filament network, including
the keratin cage(56). Thus, it is possiblethat the remainingvimentin
ifiament
cage surrounding
the spindle is more permeable to asbestos
fibers, allowing more chancesfor a deleterious chromosome-fiber
encounter.
ACKNOWLEDGMENTS
WethankE. C. Mandevilleforprovidingtheantikeratin
immunofluorescent
micrographs of newt lung epithelial cells.
chromosomebehavior.
One of the more important observationsof our study was that
crocidolite fibers did not show any affinity for chromosomes; i.e., they
did not stick to chromatin (Fig. 3). In rare occasions(2 of 146
encounters), an end of a fiber did penetrate and snag a chromosome
(Fig. 2). However, in these rare situations, the chromatin did not
adhere to the fiber and the chromosome
eventually
slipped off the
fiber. We thereforeconcludethat crocidoliteasbestosdoesnot disrupt
chromosome behavior by sticking to specific componentsof the
chromosomeor spindleapparatus.
Most chromosome-fiberencountersusually resulted in the fiber
passively yielding to the chromosome(Table 1). Moving chromo
as it squeezed around the stationary
fiber,
and in one case the chromosome was snagged and its movement was
momentarily
delayed by the fiber. Moreover,
gation (Figs. 3 and 4). From our results we hypothesizethat, for
fibers
to affect
chromosome
movement,
247: 294—301,1990.
2. Merchant,3.A. Humanepidemiology:a reviewoffiber typeandcharacteristicsin the
development
of malignantandnonmalignant
disease.Environ.HealthPerspect.,
88:
287—293,
1990.
3. Selikoff, I. J. Historical developmentsandperspectivesin inorganicfiber toxicity in
man.Environ.HealthPerspect.,88: 269-276, 1990.
4. Selikoff, I. J., Hammond,E. C., and Churg, J. Asbestosexposure,smoking, and
neoplasia.JAMA 204: 101—110,
1968.
5. Barrett, J. C., Lamb, P. W., and Wiseman,R. W. Multiple mechanismsfor the
carcinogenic
effectsof asbestos
andothermineralfibers.Environ.HealthPerspect.,
6. Craighead,J. E. Currentpathogeneticconceptsof diffuse malignantmesothelioma.
Hum.Pathol.,18: 544—557,
1987.
7. Goodglick,L A., andKane,A. B. Cytotoxicity oflong andshortcrocidoliteasbestos
fibersin vitroandin vivo.CancerRca.,50: 5153—5163,
1990.
8. Walker,C., Everitt,J., andBarrett,J. C. Possiblecellular andmolecularmechanisms
for asbestoscarcinogenicity.Am. J. Ind. Med., 21: 253-273, 1992.
9. Weisburger,
J.H., andWilliams,G. M. Carcinogen
testing:currentproblems
andnew
approaches. Science (Washington DC), 214: 401—407,1981.
10. Casey,G. Sister-chromatid
exchangeand cell kineticsin CHO-Ki cells,human
fibroblastsand lymphoblastoidcells exposedin vitro to asbestosand glass fibre.
Mutat. Res.,116: 369—377,
1983.
11. Oshimura,M., Hesterberg,T. W., Tsutsui,T., and Barrett,J. C. Correlationof
asbestos-inducedcytogeneticeffects with cell transformationof Syrian hamster
embryocellsin culture.CancerRes.,44: 5017-5022 1984.
12. Kenne, K., Ljungquist, S., and Ringertz, N. R. Effects of asbestosfibers on cell
division, cell survival, and formation of thioguanine-resistantmutantsin Chinese
hamsterovary cells. Environ. Res.,39: 448—464,
1986.
13. Oshimura,M., Hesterberg,T. W., andBarrett,J. C. An early,nonrandomkaryotypic
changein immortal Syrian hamstercell lines transformedby asbestos:trisomy of
chromosome11. CancerGenet.Cytogenet.,22: 225—237,
1986.
14. Lechner,J. F., Tokiwa,T., LaVeck,M., Benedict,W. F., Banks-Schlegel,S.,Yeager,
Jr.,H., Banerjee,A., andHarris,C. C. Asbestos-associated
chromosomal
changesin
humanmesothelialcells. Proc.Natl. Acad. Sci. USA, 82: 3884—3888,
1985.
15. Popescu,N. C., Chahinian,A. P., and DiPaolo, J. A. Nonrandomchromosome
alterationsin humanmalignantmesotheioma.CancerRes.,48: 142-147, 1988.
16. Libbus,B. L, andCraighead,
J. E. Chromosomal
translocations
withspecificbreak
pointsin asbestos-induced
rat mesotheliomas.CancerRes.,48: 6455-6461, 1988.
17. Craighead, J. E., Akley, N. J., Gould, L B., and Libbus, B. L Characteristicsof
at least one chromosome
break in our study may have been caused by an anchored fiber
snagging a chromosome arm as the spindle was rotating (Fig. 5). By
contrast, unanchored fibers, even large ones located between the
chromosomeand the pole, seemunableto affect chromosomesegre
crocidolite
1. Mossman,B. T., Bignon, J., Corn, M., Seaton,A., and Gee, J. B. L Asbestos:
scientificdevelopmentsandimplicationsfor publicpolicy. Science(WashingtonDC),
81:81—89,
1989.
somes,with an estimatedforcebetween10@ and 10_6dynes(52, 53),
easilydisplacedfibersfoundwithin the spindle.On the otherhand,the
generatedforce of a moving fiber was usuallynot enoughto displace
a chromosome.We observedonly once a moving fiber significantly
displacing a chromosome. In this particular case, a long (12.9 p.m),
thin (0.4 pin) fiber penetrated an unattached chromosome and
draggedthe entire chromosomepoleward (Fig. 2). Perhapsthe large
surface area of this extra long fiber allowed more molecular motors to
engagein force generation,and the majority of the force was directed
in one direction. Whatever the situation, the combined net force of
these molecular motors was great enough to move a large unanchored
chromosome.
Perhapsmore pertinentto the inductionof chromosomeabnormal
ities by asbestos may be the stationary fibers that do not yield to
moving chromosomes.Such fibers were found primarily along the
edgeof the spindleandarebelievedto be anchoredin the keratincage.
Unlike similar-sized fibers within the spindle, these fibers did not
yield when hit by a moving chromosome. Instead, the chromosome
usually became deformed
REFERENCES
they must be
tumorsandtumor cells culturedfrom experimentalasbestos-induced
mesotheliomas
in rats.Am. J. Pathol.,129: 448—462,
1987.
18. Harris, C. C., Lechner,J. F., Yoakum, G. H., Amstad, P., Korba, B. E., Gabrielson,
E., Grafstrom,R., Shamsuddin,A., andTrump, B. F. In vitro studiesof humanlung
carcinogenesis.
In: J. C. BarrettandR. W. Tennant(eds.),Carcinogenesis,
Vol. 9, pp.
257—269.
New York: RavenPress,1985.
19. Hesterberg,
T. W., Butterick,C. J., Oshimura,M., Brody,A. R., andBarreft,J. C.
Role of phagocytosisin Syrian hamstercell transformationand cytogeneticeffects
inducedby asbestosand short and long glassfibers. CancerRes.,46: 5795—5802,
1986.
strongly anchored to the cellular matrix. In this context long thin
fibers may be more carcinogenicthan shorterthicker ones because 20. Rieder,C. L, Sluder,G., and Brinkley, B. R. Somepossibleroutesfor asbestos
inducedaneuploidyduringmitosisin vertebratecells.In: C. C. Harris,J. F. Lechner,
they are thin enough to penetrate the intermediate filament cage
and B. R. Brinkley (eds.),Cellular and Molecular Fiber Carcinogenesis,pp. 1—26.
Cold SpringHarbor,NY: Cold SpringHarborPress,1991.
surrounding the forming spindle and long enough to extend well into
21. Hesterberg,T. W., and Barrett, J. C. Induction by asbestosfibers of anaphase
the spindlewhile remaining anchoredwithin the cage.
abnormalities:mechanismfor aneuploidyinduction and possibly carcinogenesis.
Mesothelial cells are 10 times more sensitivethan bronchial epi
Carcinogenesis(Land.), 6: 473—475,
1985.
thelial cells to asbestos(14, 18). An important difference between 22. Palekar,L D., Eyre,J. F., Most, B. M., andCoffin, D. L Metaphaseandanaphase
mesothelialandepithelial cells is their intermediatefilament network
(54, 55). Mesothelial cells have different keratin types and produce
morevimentin thando epithelialcells(54). Mesothelialcellsnormally
have a high keratin/low vimentin content in vivo, but during cell
division
it appears that the keratin
content drops and the vimentin
content rises (55). Mesothelial cells round during cell division and,
like certain immortalized epithelial cell lines, this rounding is proba
analysisof V79 cells exposedto erionite, UICC chrysolite and UICC crocidolite.
Carcinogenesis(Lond.), 8: 553—560,
1987.
23. Yegles, M., Saint-Etienne, L, Renier, A., Janson, X., and Jaurand, M. C. Induction of
metaphaseand anaphase/telophase
abnormalitiesby asbestosfibers in rat pleural
mesothelialcells in vitro. Am. J. Respir.Cell Mol. Biol., 9: 186—191,
1993.
24. Wang,N. S.,Jaurand,M. C., Magne,L, Kheuang,L, Pinchon,M. C., andBignon,
J. The interactionsbetweenasbestosfibersandmetaphasechromosomesof ratpleural
mesothelialcells in culture.Am. J. Pathol.,126: 343—349,
1987.
25. Pots, F. Problems in defining carcinogenic fibers. Ann. Occup. Hyg., 31: 799—802,
1987.
797
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
CROCIDOLITE
A5BESTO5AND MITO5IS
26. Stanton, M. F., Layard, M., Tegeris, A., Miller, E., May, M., Morgan, E., and Smith,
light and electron microscopic studies on the mechanisms of mitosis in vertebrate
somatic cells. Proceedings of the 13th International
Congress on Electron
A. Relation of particle dimension to carcinogenicityin amphiboleasbestoses and
other fibrous minerals. J. Natl. Cancer Inst., 67: 965—975,1981.
27. Lippmann, M. Effects of fiber characteristics on lung deposition,
Microscopy, 3: 75—78,1994.
41. Bruch, J. Response of cell cultures to asbestos fibers. Environ. Health Perspect., 9:
253—254,
1974.
42. Johnson, N. F., and Davies, R. The effect of crocidolite and chrysotile on peritoneal
retention, and
disease. Environ. Health Perspect., 88: 311—317,1990.
28. Stanton, M. F., and Wrench, C. Mechanisms of mesothelioma induction with asbestos
and fibrous glass. J. Natl. Cancer Inst., 48: 797—821,1972.
macrophages: a study by transmission and scanning electron microscopy. In: R. C.
Brown,M.Chamberlain,R. Davies,andI. P. Gormley(eds.),Thein vitroeffectsof
29. Hesterberg, T. W., and Barrett, J. C. Dependence of asbestos- and mineral dust
mineral dusts, pp. 97—103.
New York: Academic Press, 1980.
induced transformation of mammalian cells in culture on fiber dimension. Cancer
Res., 44: 2170—2180,1984.
30. Cole, R. W., Ault, J. G., Hayden, J. H., and Rieder, C. L. Crocidolite asbestos fibers
43. Miller, K. The effects of asbestos on macrophages. CRC Crit. Rev. Toxicol. 5:
319—354,1978.
undergo size-dependent microtubule-mediated transport after endocytosis in verte
44. Huang, S. L., Saggioro, D., Michelmann, H., and MaIling, H. V. Genetic effects of
crocidolite asbestos in Chinese hamster lung cells. Mutat. Rca., 57: 225—232,1978.
45. Hayden, J. H., Allen, R. D., and Goldman, R. D. Cytoplasmic transport in keratocytes:
direct visualization of particle translocation along microtubules. Cell Motil., 3: 1—19,
brate lung epithelial cells. Cancer Res., 51: 4942—4947,1991.
31. Jensen, C. G., Jensen, L. C. W., Ault, J. G., Osorio, G., and Rieder, C. L. Time-lapse
video light microscopic and electron microscopic observations of vertebrate epithelial
cells exposed to crocidolite asbestos. in: J. M. G. Davis and M-C. Jaurand (eds.),
Cellular and Molecular Effects of Mineral and Synthetic Dusts and Fibres, NATO
1983.
46. Vale, R. D., Scholey, J. M., and Sheets, M. P. Kinesin: possible biological roles for
a new microtubule motor. Trends Biol. Sd., 11: 464—468,1986.
47. Gibbons, I. R. Dynein A1'Pases as microtubule motors. J. Biol. Chem., 263: 15837—
15840, 1988.
ASISeries,Vol.H 85; pp. 63—78.
Berlin:Springer-Verlag,
1994.
32. Hesterberg, T. W., Brody, A. R., Oshimura, M., and Barrett, J. C. Asbestos and silica
induce morphological transformation of mammalian cells in culture: A possible
mechanism. in: D. F. Goldsmith, D. M. Winn, and C. M. Shy (eds.), Silica, Silicosis,
and Cancer, pp. 177—190.New York: Praeger Press, 1986.
48. McIntosh, J. R., and Porter, M. E. Enzymes for microtubule-dependent motility.
J. Biol.Chem.,264:6001—6004,
1989.
33. Jaurand,M. C., Bastie-Sigeac,
I., Bignon,J.,Stoebner,
P., Renier,A., andKheuang, 49. Vale, R. D. Microtubule-based motor proteins. Curt. Opin. Cell Biol., 2: 215—222,
L. Effectof chrysotileandcrocidoliteon the morphologyandgrowthof rat pleural
1990.
mesothelial cells. Environ. Rca., 30: 255—269,1983.
34. Davies, R., Allison, A. C., Ackerman, J., Butterfield, A., and Williams, S. Asbestos
induces selective release of lysosomal enzymes from mononuclear phagocytes. Na
ture (Land.), 25!: 423—425,1974.
35. Rieder, C. L., and Hard, R. Newt lung epithelial cells: cultivation, use, and advantages
for biomedical research. Int. Rev. Cytol., 122: 153—220,1990.
50. Vale, R. D., and Goldstein, L. S. B. One motor, many tails: an expanding repertoire
of force-generating enzymes. Cell, 60: 883—885,1990.
51. Waters, J. C., Cole, R. W., and Rieder, C. L. The force-producing mechanism for
centrosome separation during spindle formation in vertebrates is intrinsic to each
aster. J. Cell Biol., 122: 361—372,1993.
52.
properties of asters: astral microtubule turnover influences the oscillatory behavior
and positioning of mono-oriented chromosomes. J. Cell Sci., 99: 701—710,1991.
R. B. The forces
that move
chromosomes
in mitosis.
Ann.
Rev.
Biophys.
53. Alexander, S. P., and Rieder, C. L. Chromosome motion during attachment to the
vertebrate spindle: initial saltatory-like behavior of chromosomes and quantitative
37. Rieder, C. L., and Alexander, S. P. Kinetochores are transported poleward along a
single astral microtubule during chromosome attachment to the spindle in newt lung
cells. J. Cell Biol., 110: 81—95,1990.
38. Mandeville, E. C., and Rieder, C. L Keratin filaments restrict organelle migration
into the forming spindle of newt pneumocytes. Cell Motil. Cytoskeleton, 15:
111—120,
1990.
39. Rieder, C. L, Davison, E. A., Jensen, L C. W., Cassimeris, L, and Salmon, E. D.
Oscillatory movements of monooriented chromosomes and their position relative to
analysisof force productionby nascentkinetochorefibers. J. Cell BioL, 113:
805—815, 1991.
54. Wu, Y.-J., Parker, L M., Binder, N. E., Beckett, M. A., Sinard, J. H., Griffiths, C. T.,
and Rheinwald, J. G. The mesotheial keratins: a new family of cytoskeletal proteins
identified in cultured mesothelial cells and nonkeratinizingepithelia. Cell 31:
693—703,1982.
55. Connell, N. D., and Rheinwald, J. G. Regulation of the cytoskeleton in mesothelial
cells: reversible loss of keratinand increase in vimentin during rapid growth in
culture.Cell34: 245—253,
1983.
the spindle pole result from the ejection properties ofthe aster and half-spindle. J. Cell
Biol.,103: 581—591,
1986.
Nicklas,
Chem.,17: 431—449,
1988.
36. Ault, J. 0., DeMarco, A. J., Salmon, E. D., and Rieder, C. L. Studies on the ejection
56. Lane, E. B., Goodman, S. L, Trejdosiewicz,
L K. Disruption of the keratin filament
network during epithelial cell division. EMBO J., 1: 1365—1372,
1982.
40. Rieder, C. L., Schultz, A., Cole, R., Ault, J., Waters, J., and Salmon, E. D. Correlative
798
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
Behavior of Crocidolite Asbestos during Mitosis in Living
Vertebrate Lung Epithelial Cells
Jeffrey G. Ault, Richard W. Cole, Cynthia G. Jensen, et al.
Cancer Res 1995;55:792-798.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/55/4/792
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.