0022-15M/79/2701-0014$02.OO/O THE JOURNAL OF HISTOCHEMISTRY

Copyright

© 1979 by The

AND

Vol.

CYTOCHEMISTRY

Histochemical

Society,

Sedimentation

Velocity

Separation: A Preparation Cervical Samples1

K. OTTO, Institut

f#{252}r Klinische

Zytologie

der

H. HOFFKEN TU

M#{252}nchen,

Received

H.-J.

AND

M#{252}nchen 80, Federal

8’XX’

June

The

need

for

developing

directed

to

this

Basically,

goal

(3,

all these

ent automated signals only. tional those

new

preparative

in cytology 6, 7, 9,

efforts

analytical Looking at

Papanicolaou signals (1, 4).

Overlapping

has 15,

smear

from

26,

limitation

reveals

only

small

of pres-

proportions

Two (a)

tors.

borders,

leukocytes

and to a certain cause false alarms

thus

intolerable

to

degree overlapping in those systems

increase

of false

of residual

immediate

immersion

30 parts

of 96%

Preparation on a Vortex-mixer off the applicator.

nuclei (2) and

positive

for

Republic

of Germany

exfoliative material on glass features of this procedure are over isopycmc medium of 1.026 50 x g yields two preparations on the first slide and epithelial as the amount of diagnostically resolution analysis.

from

the uterine cervix are prepared of a conventional Pananicolaou material of

in

the

10 ml

applicator

of suspension into

suspension

from smear

each and

medium;

(b)

medium.

Figure 1 shows the components used for sampling and mailing. Polyropylene suspension tubes are used to prevent mechanical damage of samples. The suspension medium consists of 70 parts phosphate buffered salt solution (PBS) with Ca and Mg omitted

of

superimposed

samples preparation

immersion

and

cellular

to an

the

14-18, 1979 in U.S.A.

1978

case:

which can process separate cells as signals, the conven-

epithelial cells would therefore lead

for

of efforts

16).

arise

systems isolated

procedures

led to a variety

Method

1, pp. Printed

SOOST

A preparation procedure, aiming at monolayer deposition of cervical slides for high resolution prescreening has been developed. The main centrifugal depositon after suspension and sedimentation of samples density. Fractioning of the separation column after centrifugation at with leukocytes, bacteria and cellular debris predominantly located cells on the second one. The degree of spatial cellular isolation as well relevant cells per slide seem to fit the requirements of automated high

prescreening-systems

27, No.

Inc.

ethanol

(9).

procedure: Suspended samples are shaken 30 sec (Pos. 3) during which the cottonswab readily slips Cotton is removed from suspension by filtration

machine

decisions.

Isolation gether

of exfoliated with

such

removal

of

as leukocytes,

enrichment

Fluorescence

promising

starting

Double presented vides

to provide

staining here

less

of samples a Papanicolaou controls the

of

System for is collected I

Supported

nologie,

Bonn.

noise”

mucus

sorting

etc.,

machines

for

be avoided. stain is used examination

by

Gynecologic cotton-tipped

Bundesministerium

Germany,

analysis. .-

In the procedure (13), which protrained

U’

cyto-

results.

specimen collection: by gynecologists using the

0

if an absorp-

subsequent

METhODS

West

is a

enriched

means of sedimentagrandient centrifu-

AND

by

U’-

to these

(FACS)

procedures,

is used

for

are

suited

(1 1, 14).

complex can

and

cells

highly

samples

cells

procedure

immediate

cell

0-

to-

“biologic relevant

constituents by (5, 12) or density

MATERIALS rial

and excess

producing

relevant

staining

pathologists

sheets masses,

activated

of sample elutnation

seems

tion-based

a suspension

cytodiagnostically

point,

diagnostically

gation

cellular

preparing

to be met in a preparation procedure analysis. in several directions has been applied

problems.

Separation tion (3, 16),

by

bacterial

of isolated,

the requirements for automated Recent work

with

cells

Grants

C) exfoliative plastic

f#{252}r Forschung

RV12/GfW/2

und

and

mateapplica-

3

Tech-

FIG.

01VH127/

Applicator;

ZA/NT/MT/225a.

1. System

for

2, suspension

sampling

tube;

14

Downloaded from jhc.sagepub.com at UNIV OF ILLINOIS LAW LIBRARY on March 10, 2015

and

mailing

3, mailing

of suspended

box.

material.

1,

SEDIMENTATION through

a nylonnet

estimate

of

total

suspension layered

in on

20#{176}C.The

80-s

content

a counting

top

porewidth

of the

Homburg,

(Nytalmonofilament).

is performed chamber.

of 8 cm1

Bad

Fresenius, at

with cell

by counting

About

West

Germany)

with

component

of

aliquot

are (Fa.

1.026 g/cm’

After

of

Plasmasteril

this

in 0.9%

and

An an

of suspension

medium

hydroxyethyl)-amylopectin-hydrolysate

Before

5 cm

separation

macromolecular

VELOCITY

density

medium

is

NaCl-solution.

O-(2After

Sedimentation

Sample

proximal

--.1.Separation

I

Column

I

-Fractioning

-

distal

centrifugation at 50 x g for 10 mm, leukocytes, bacteria and cell debris are found in the proximal part of the separation column as related to the rotor axis and epithelial cells are found in the distal part. The geometry of the glass tube used for cell separation is as follows: total length: 10.8 cm, inner diameter: 1.4 cm, length of separation column: 5.2 cm, length of supernatant cell suspension: 3.1 cm. The column is fractionated in two parts as shown in Figure 2, by means of pipetting off 5.5 cm. Thus, preparation 1 consists of supernatant and a small proximal part of the separation column. This preparation is centrifuged on a separate slide by means of newly developed centrifugal buckets (9, 10). Figure 3 gives different views of these buckets. They consist of a polypropylene bottom and pressure plate with polyisoprene-deposition chambers pressed to the slides by means of two clips. In preparation 2 (distal part of column) cell concentration is again estimated as described above and dilution to about 10.000 epithelial cells per cubic centimeter takes place if necessary. Of this solution 5 cm’ (up to 50,000 epithelial cells) are centrifuged onto a coated glass slide with 1100 x g for 20 mm. Slides are coated (5, 8) before mounting with 1 drop of 0.1% aqueous solution of poly-L-lysin (MW -70.000, Fa. Serva, Heidelberg) by spreading this drop on the glass surface with the side of a Pasteur-pipette. Fast evaporation of water is achieved by heating the slides. After centrifugation supernatant is pumped out of the chamber. This

FIG.

gation

2. Fractioning 50 x g. Proximal

of separation

column

and distal

10 mm

after

fraction

are prepared

separately.

Volume

ofone

is followed

chamber.

of centrifu-

15

SEPARATION

After

if pumped

by

addition

10 mm

off and

of 5 cm

of repeated

chambers

are

of 96%

clipped

chamber:”..6ccm

Chambe

Holder

for

Pressure

Bottom

Top

a FIG.

-

2 3.

Construction

pressure

plate

plate

with

slides

view

3

4

cm of centrifugal

buckets

for

cell

deposition

on

glass

Downloaded from jhc.sagepub.com at UNIV OF ILLINOIS LAW LIBRARY on March 10, 2015

ethanol

centrifugation

slides.

off the

to the at 1100

slide.

mounted

x g fixative

16

Orro

No

epithelial

cells

have

been

found

in

supernatants

of

cases. After abother 20 min of fixation in 96% ethanol directly transferred to the Papanicolaou-staining-machine Varistain 23) without intermediate air-drying.

the

ET

various

slides

superimposed

are

(Shandon

FIG.

cellular

4. Microscopic

appearance

superimpositions.

of control

slide

stain,

(xlO).

Papaniclaou

r.

-:a--

#_

1 C

..

..

: ‘r.,, 4#{149}

.

the

S

.4,

S

the been

extremely

separation drastically

high

rates

#{149}

.

.A

C

.

,

.v

#{149}. S

..

#{149}5

#{149}

a

S

.

s #{149}

S

#{149}:

r’

-#{149}:iii’’h

...

-II.

S

.#{149}. #{149}#{149}

1cst.:

;-‘

#{149}0#{149} .

e#{149}

, .

#{149}‘-

.

-

a

.5,

S

. S #{149}#{149},e :‘

#{149} ,.#{149}.-

#{149}

‘3_.,S

-

0C

%.:#{149}‘

#{149} 5(5

??

.

‘

4#{149}

S

.

..

S.

#{149}

I.

#{149}*A.

e

.

.

-#{149}

.1*.#{149}Z#{149}

-*‘#{149} .,#{149}.

a

l:.

!#{149}

41#{149} S

C ‘

.

5.#{149}.

.

dIi,

C #{149}5#{149}C .5

5. Microscopic Enrichment

Note

of has

of 6 is a and

.

#{149}‘1

FIG.

column. Figure after sedimentation part

case

deposition

.

‘

.

column.

centrifugal

same

S.-,.

#

4#{149}/ #{149}5

Pananico-

of the

#{149}.

.

#{149}.

a slide

the distal of leukocytes

cervix.

#{149}#{149}-

.

-1

#{149}

of

IS

C

aA

.

the separation same case,

uterine

#{149}#{149}

,.

!S

and

deposition proportion

S

#{149}

sedimentation of the

of the

in a conventional

5 shows

part from

The

‘.5.

#{149}

cells

Figure

a

.

,

,-$;I’#-

-a

suspension,

,

,Y

S

whereas

carcinoma

.,

Iw.r.

.t

after

with

S

:;

smear,

centrifugal column.

a patient

#{149}

1

#.

‘. .q

is demonof leukocytes

from

S.

a

,

morphology masses

epithelial

on

laou

the proximal preparation

RESULTS Microscopic appearance of cellular strated in Figures 4-6. Figure 4 exhibits

AL.

appearance of leukocytes.

I’.

5

of monolayer Papanicolaou

.

S

preparation stain, (x 10).

after

sedimentation,

same

-

case

I’

!

-

as in Figure

Downloaded from jhc.sagepub.com at UNIV OF ILLINOIS LAW LIBRARY on March 10, 2015

4. Proximal

part

of

separation

of

SEDIMENTATION

VELOCITY

SEPARATION

17

.

Ic

S ‘p

I

a

S

I %4 C

4

S.

I.”, I’

a

S

I ‘Si

a

Removal

6. Microscopic of leukocytes,

reduced,

epithelial

FIG.

the

cells

whole field The background

lular

overlap

appearance reduced

of monolayer preparation after cellular overlap-rates. Papanicolaou

are

of view. is clear,

located

in one

intracellular

is relatively

high

plane

contrast

in this

because cell concentration had cells to be deposited on 6 cm2.

focal

been

across

is high.

particular

sedimentation, same stain, (X 10).

Cel-

preparation,

adjusted

to a total

of i0

separation

procedure particles

is based in suspension

ences. Assuming cells

effect

very

and

white

of the

on

high

cellular

elements

we use

different sedimentation rather than on particle particle

blood

densities

cells

(>1.06),

in our

velocities of density differ-

cells.

The

for

both

epithelial

sedimentation

distal

part

sedimentation

question

of the

medium,

are pipetted technique

we

off after found

remains,

is less

sedimentation. substantial

in the

cases, mized.

but loss of diagnostically Until now, about 150

including noma in uterine

separation In three

distal

separation which

cated

part

whether

There procedure

of the

samples

were failed. of cervical

column, than

two

Fig. 2) By this of leukocytes lo-

separation

medium

relevant samples

cells have

samples, proven types

carcinoma

in various

can be minibeen prepared,

dysplasias, carcinoma

of samples with

carciof the where

extremely

our high

as in Figure

infiltration

respectively. Therefore, cases.

suspension of the

our

results

mm,

50

medium

(1.026

samples

for

or

g)

g/cm3)

i. a valuable

automated

high

covermasses

smears in cases

that

of gynecologic

the very cases on

as well

is necessary

suspension

conclude,

formed to block

of complete bacterial

conventional

masses was used described before. we

x

barrier

inflammation smears of these

preparations

original

leukocytic blocking

(10

of the

derived

column.

medium

were not readable because cells by granulocytes

Dilution

From tion

of separation

separation

bacterial Control

a repetition

as of the

part

a mechanical

of the

two cases of massive effect was observed. other hand of epithelial

4. Distal

we found

leukocytes on top cell sedimentation.

undifferen-

only 0.5 cm3 of 0.5 cm in depth,

(compare proportions

gynecologically normal situ and histologically

cervix.

In same

velocity

tiated epithelial cells (e.g., parabasal cells) or small tumor cells can be separated from leukocytes under experimental conditions described here. In order to collect those cell types in the

by the epithelial

rial and/or sedimentation

differences depend on the square of the diameter of spherical particles, according to Stokes’ law. Thus, bacteria, cellular debris and small leukocytes can easily be separated from large epithelial

granulocytic

the ing

DISCUSSION The

case

to avoid short

resolution

with the

time

samples tool

in those

in an

in preparing

bacteeffect

of

sedimentsisopycnic adequate

analysis.

ACKNOWLEDGMENTS The

technical

assistance

of Ms.

U. Schafer

and

Mrs.

M. Eifler

is

acknowledged. LITERATURE

CITED

Bibbo M, Bartels PH, Chen M, Harris HJ, Truttmann B, Wied GL: The numerical composition of cellular samples from the female reproductive tract, The Automation of Uterine Cancer Cytology, Tutorials of Cytology, Chicago, Illinois, 1976 2. Cambier MA, Wheeless LL, Patten SF: False alarms: current obstacle to cytopathology automation. Acta Cytol 20:586, 1976 3. Danno, M, Furuyama Y, Kusama S, Sakamaki A, Nosaka K, Kashida R: A cell preparation method using liquid media for pattern recognition of automated prescreening systems. Proceedings of the Second International Conference on Automation of Cancer Cytology and Cell Image Analysis, Tokyo, 1977 4. Feldman, RT, Poulsen R, Shepherd L, Marshall KG: The occurance of isolated dysplastic and carcinoma in situ-type cells in 1.

Downloaded from jhc.sagepub.com at UNIV OF ILLINOIS LAW LIBRARY on March 10, 2015

OTTO

18

cervical smears from patients with dysplasia and carcinoma in situ: Significance to prescreeing using image processing techniques. Acta Cytol 17:395, 1973 5. Grabske RI, Lake S, Gledhill BC, Meistrich ML: Centrifugal elutriation: Separation of spermatogenic cells on the basis of sedimentation velocity. J Cell Physiol 86:177, 1975 6. Husain OAN, Allen RWB, Hawkins EJ, Taylor JE: The quantimet cytoscreen and the interactive approach to cancer screening. J Histochem Cytochem 22:678, 1974 7. Leif, RC, Gall S. Dunlap LA, Railey C, Leif SB: Zucker RM: Centrifugal cytology IV. The preparation of fixed stained dispersions of gynecological cells. Acta Cytol 19:159, 1975 8.

Mazia

D, Schatten

G, Sale

W: Adhesion ofcells Biol 66:198, 1975

to surfaces

coated

with polylysine. J Cell 9. Otto K, Bayer-Pietsch E, Dvorak R, Soost H.-J.: Ein Verfahren zur isolierten Deposition von Zellen der Cervix uteri auf Glasobjekttr#{228}ger. Biotechnische Umschau 2:128, 1978 10. Otto K: Automatische Analyse von Zellen der Cervix uteri unter Verwendung konventioneller und neuer Pr#{228}parationsverfahren.

ET

AL.

11.

Ph.D.

dissertation,

Pulse

Cytophotometry. Cytophotometry”,

The

Technical

University

Proc Ilird Vienne

International Austria, 1977

“Pulse Gent. In press

of Munich, Symposium European

1978 on Press,

H, Soost HJ: Grundri$ und Atlas der gyn#{224}kologischen Zytodiagnostik, 3. Auflage, Thieme-Verlag, Stuttgart, 1971 13. St#{246}hr M, Goerttler K: Dual laser flow analysis and sorting of cells including the uterine cervix. IInd International Conference on Automation of Cancer Cytology and Cell Image Analysis, Tokyo, Japan, 1977 14. Tanaka N, Ikeda H, Ueno T, Watanabe 5, Imasata Y, Tsunekawa 12.

Smolka

S:

Fundamental

study

of

automated

cytoscreening

for

uterine

cancer. IV. Sample requirements for cybest and simulation test of cell dispersion. Acta Cytol 21:531, 1977 15. Tolles WE: Garcia GL: Enrichment of the gynecologic cytologic specimen. J Histochem Cytochem 25:504, 1977 16.

Tucker

JH,

for automatic 947, 1971

Gresham

GA:

screening.

Downloaded from jhc.sagepub.com at UNIV OF ILLINOIS LAW LIBRARY on March 10, 2015

Preparation

J Obstet

of cervical

Gynecol

scrape

Br Commonwealth

material

78: