1554/79/2701-0193$02.00/0

0022THE

JOURNAL

Copyright

OF HISTOCHEMISTRY

© 1979 by The

AND

Vol. 27, No. 1, pp. 193-198, 1979 Printed in U.S.A.

CYTOCHEMISTRY

Histochemical

Society,

Transformation

Inc.

of Nuclear

Morphology

J. M. TEICH,2 Research

Laboratory

of Electronics,

I. T. YOUNG,

Massachusetts Received

during S. E. SHER

Institute for

AND

Cellular J. S. LEE

of Technology,

publication

June

20,

Maturatio&

Cambridge,

Massachusetts

02139

1978

Using in vivo labeling of mouse leukocytes with tritiated thymidine, cells in the neutrophilic series were studied to determine the change in their nuclear shape as a function of maturation level. Several morphologic shape parameters including perimeter and bending energy were used to quantify the distribution of the nuclear morphology in a given age cohort. The change in these distributions as a function

of calendar

possibility

In

our

we were related (9),

earlier

publications

level

the

on

concerned with to a biophysical

a comparison

age

of inferring

the

shape

The

two

quantitative

of biologic

with

I t.obeci

I My&ocytee

:hiIIing

flyelocytes

other

parameters

shape

problem basic clear

proposed

were

used

to

test

the

counting

(5). In this

paper

we address

concerning the relationship of a cell and its known

here

between

calendar

between calendar

age

(relative

a more the age.

to

nuWe

mitosis)

l1atupe_Neutpophils eLblclalblcId1JIcJjI.fRTdk]F

I5tbForm

I

U

I Lobe

rasy

above

I g LobesW l?ji!sci

Ponder

of lobe problem morphology

distinguish

Nucleus

IJuvenhIe

:ooseerld

named

morphology.

Neutr’opriils

Imm6.tuPe

Aj,ieth

ft

the

measure of shape that of the cellular membrane

measure

bs,fcaZlsi

determined.

from

analysis

a quantitative property

of that

was

age

#{244}eqmented m I t

I

I

2LobesI5Lobesl4LobeI

5or1oreLoboD ---.-.-.-

(I)

Ais

No,i-F,Iernerrted

(2)

Fil8rneri

ted

‘3

‘;:#{176} 50 U,I.,

t’,0

.

6’

o

,1,to9 ;

0 A

Cell Type ; MyobsttPrompoMyelocyte

I cY

r1etamyel

e

#{182}ilem’nt

cyte

--;-

2 Lobes

utmphl

r’uerteci

Ncutroph.I’

Norm_g..._. Extreme

5Mift

Moderate 1. Conventional

FIG.

“Principles measures and

the 1

2

and

with

Supported

in

and

descriptions

in Wintrobe,

human

incorporation

Supported,

AG00354.

labels

of Hematology”,

assessment

of syntactic part,

by a Hertz

by the

National

Foundation

of the

Clinical of shape procedures Institute Fellowship.

maturation

Hematology,

the

on Aging

..IFTt

sequence

a,

for

Lea & Febiger,

complexity for

-._____ to

5hiFV

To Left’

(4) specific Grant

and

functional

as tures,

mations

age, It is well

a human its

- -

-

Left’

human neutrophiic Philadelphia, 1967).

functions.

.

Right

i.e.,

leukocytes.

the

established

polymorphonuclear

nucleus undergoes from an essentially

193

Downloaded from jhc.sagepub.com at UNIV OF SASKATCHEWAN on March 18, 2015

(Taken

ability

to

in the

from

perform literature

leukocyte

Haden,

cell

R.

L.,

specific

(3, 6, 7) that (neutrophil)

ma-

a series of morphologic transforround shape to an indented shape

194

ET.

TEICH,

(band

form)

This

type

have

and

been

attached

illustrated

in

To

study

turned

fmally to

Figure

this

(segmented)

as well

various

labeling the

the central opens the

as the

different

stages

form. labels

by

eters that describe nucleus maturation

that

authors,

a potential

is

cell in a truly

model

system.

nucleus

form to an open 2.) To determine quantify

the

phenomenon

of the

pulse

a multiobed

1.

to a mouse

mation

to

of transformation,

shape

ring, the

cells

in vivo

of the

cells

ring-indicative value of this

measure is minimized outer boundary-measurement cific

boundary

in terms

that

boundary

out

with we

useful

for

of the

(c) bending of the complexity (work)

thin-shelled

within

ratio.

squared This

energy of the of the spe-

required

to form

medium

(see

Ref.

ofthe nucleus; (e) perimeter ofthe outer boundary. Statistical procedures were then used to summarize the nuclear shapes as a function of age, determine the distribution as a function

of age

and

determine

the

of immature

for

neutrophils

MATERIALS

9); (‘0 area

of shapes

and

stages would

predicting

cell

estimating,

for

in a blood

in the cell then offer age

from

example,

the

sample.

key

AND

METHODS

Four mice received injections with a pulse of tritiated thymidine (5 .iCi/g) and four were used as controls. Six hours after the initial injection the four test mice were injected with x 100 concentrated, unlabeled thymidine to dilute the effects of the labeled thymidine. Cells synthesizing DNA in the bone marrow during the 6-hr pulse period would thus incorporate the label. All eight mice received a dose of 0.5 ml unlabeled thymidine 18 hr later. Blood samples were taken from each mouse according to the schedule in Table I. Slides were prepared from the blood samples, coated with autoradiographic emulsion (Kodak Type NTB2), and exposed for a period of 2 weeks. The slides were then stained with Giemsa and examined at x400 to locate labeled cells. When a labeled cell was detected (three or more grains seen over the nucleus and a low background count), the following information was catalogued: (a) age of the cell-calculated

To

area

to the perimeter a “thinness”

the various last analysis

for objectively

morphology,

ring

(see Fig. we used

thymidine. (a)

mechanism

nucleus

fraction

and identify cycle. This

transfor-

closed form study

tritiated

we

ring form. When the ring drops to zero; (b) P2A of

energy

of a linear,

the

initially

measured:

as

a circle;

way

mouse

an

(segmented) cells under

of the parameter

the outer boundary-proportional divided by the area and

quantitative

In the

is from

multiobed age of the

of the shapes

AL.

to within

±3 hr by subtracting

the

time

plus 3 hr from the time of bleeding Y coordinates on the slide-measured

param-

oftritiated

thymidine

injection

for cells on that slide; (b) X and by using 10 steps to move the

MOUSE

:

CELLS FIRST SEEN PERIPHERAL BLOOD AS A “RING FORM” FIG.

IN

2. Stages

RING OPENS A BICORNATE

in polymorphonuclear

TO GIVE SHAPE

leukocyte

LOBES DEVELOP; NUCLEUS BECOMES IRREGULAR C POLYMORPHONUCLEAR)

maturation

in the

mouse.

© Day

1.5

Day

2.5

Day

Day

4

16

c9 Day

FIG.

columns:

3.

Examples

Day.

Day

3

of nuclear

day 4, 6, 10. Third

shape:

two

pair of columns:

randomly

day

selected

labeled

6

Day

23

Day

31

10

cells

per

age

cohort.

First

pair

16, 23, 31.

Downloaded from jhc.sagepub.com at UNIV OF SASKATCHEWAN on March 18, 2015

of columns:

day

1 .5, 2.5,

3. Second

pair

of

.

S.

.5-4

11

3

.

41 3-1

21I

1

..

3.

#{149}#{149}.

2

\1j\t\ /\f/\A

! \/

-

n..

: \R[R

L

f#{149} 21I....

.

-

.

Ii,

S

288

‘:

0

:.i

.

[\

lO

(a)

a -‘

.

i:

10 -‘

.

:j

.

8

6

.

,..

‘

.

.

I!

:

1.cco

3.C.oO

1.808

5.IeO

M

3.800

‘A

______

R

.888

5.080

3.888

5.U8

1.868

3.584

5.CO

(b) II

I,

U

S

6

.

.

(c)

. .

2888

458

1485

18_i

Day

S

1.5

Histograms of five different region; (b) P2A of the outer boundary; (e) perimeter of the outer boundary. FIG.

4.

7V\ 1408

2#{243}

(d) I,

.

‘S

2885

.

#{149}

#{149}

.

1

‘28

Day nuclear parameters (c) bending energy

3

(e)

Day

10]

S

4

/A3 Day

0

7

as a function of age (25 cells per histogram). (a) Area of central pale of the outer boundary; (d) total area of the cell, exclusive of the central 195

Downloaded from jhc.sagepub.com at UNIV OF SASKATCHEWAN on March 18, 2015

(nonnuclear) pale region;

alt.

,

8(I

S

5I

‘

i

iI

i

I

‘V1J1N

7. 11(1

Ro9E$

4.28(1

28411

I.

428$

0. 00 ii-----I-------I------I----I 0.00

3.00(3

$5081

200(1

2.H(I

I

4.008-I

I

0.00

_:,__

--I

580

eoa-i

II----I--------I--------$--------I 0.00 500

300(1

-.---

II

ll

20!tI

-..I

-

Z50(I

300(1

(b)

(a) S C. 2(3

I

3 4281

iI\ 2. 8281

a

44(1

1 34(3 18281

I 01(3

1. 47(1

,.

a.

77

71(2

4. 59

0 00

.

;,______.

------I--000

500

--

...-.-.-,-...-.-,-,-.-#{149}---.-#{149}-#{149}-#{149}-*-s-#{149}--#{149}-#{149}.0.00 .. l-------I-------I--------I----I-100(1 $EI

(c)

d.s 00(I

230(2

II---------l---------I 0.00 5.00

30011

--,

$0001

1.50(1

2!.Oul

I--3.00(1

(d)

IPsO,) .1

29E2

Si

I------.---I 2(3(1

31(2

1.9282

1.54(2

1. 1282

7.70(1

I

38281

I

0.00

aoo

11

1.iOFl

dss

--I-------I#{149}---

-I 000

20011

2i0(1

300(1

(e) FIG. 5. Plot of average nuclear shape parameters as a function of age. Brackets indicate one standard Area of central pale (nonnuclear) region; (b) P2A of the outer boundary; (c) bending energy of the outer exclusive of the central pale region; (e) perimeter of the outer boundary.

196

Downloaded from jhc.sagepub.com at UNIV OF SASKATCHEWAN on March 18, 2015

deviation boundary;

of the mean (d)

total

(N area

=

50). (a)

of the

cell,

TRANSFORMATION

TABLE Bleeding Slide

Times

for the Mice

No.

Mouse

OF

NUCLEAR

197

MORPHOLOGY

I Used

in this

Experiment” Days Original

No.

Since Injection 320 0

115-1

LSA

(A-E) 2

(5)

3

2

OoL

4

(6)

1.5

5

3

2.5

(7)

2.5

7

4

2.5

8

(8)

2.5

6

13

3

3 3 3 3 4

14

4

4

15

(7)

4

16

(8)

4

18

2

19 21

(5) (6) 3

5 5 5

22

4

23

(7)

24

(8)

9 10

2

11

(5)

12

(6)

17

20

34

2

35

(5)

10

36

(6)

37

10 16

39

3 4 (7)

40

(8)

25 26

2

27

31

(5) (6) 3 4 (7)

32

(8)

28 29 30

33

38

2

43

(5) (6) 3

31 31

47

4 (7)

48

(8)

31

U,

a-

51 52

10

0t

160

Leukocyte

nuclei

0 Data 0003C

E

0014C

Data

0

60 Bending energy 0.B.

6. Two-dimensional scattergram for the nuclear outer boundary. circle are cells of age 1.5 days; points are cells of age 4.0 days. FIG.

ergy

stage

in either

direction

and

recording

of perimeter vs bending enPoints marked with an open marked with an open triangle

the

number

of steps

to reach

a particular cell from an origin on the slide; (c) number of grains on the surface; (d) sketch of the nucleus outline, and; (e) sketch of the microscope field in the neighborhood of the located, labeled cell. Two slides in each age group were catalogued for 25 cells per slide; then the autoradiographic emulsion was removed and the slide restained with brilliant cresyl blue to maximize the nucleic contrast (2). Using the information described above the labeled cell was relocated and analysed under x 1570 magnification (6.7 picture elements per micron) using the ACUity system (8). This corrected optical system, with Kohier illumination and a 10-nm bandwidth interference filter at center wavelength 570 nm uses a Vidicon scanner and a 6-bit A/D converter to represent each brightness point in the image from black (brightness = 0) to white (brightness = 63). Field shading throughout the field of observation corresponded to less than 1 part in 64, hence correction was not required. A brightness threshold, corresponding to the minimum between the maxima of the two darkest modes of the trimodal (nuclei, cytoplasm and background) brightness histogram, was calculated and used to separate the nucleus (dark material) from the cytoplasm and background (light material). The boundary of the nucleus was then extracted and used to compute the various shape parameters. Each nuclear boundary was displayed and examined to verify that it corresponded to the complete boundary observed through the microscope. Minor alterations in the threshold were made when necessary.

38

2 (5) (6)

50

0

&

E

31

49

0

U,

16

42

0

0,

16 23 23 23 23

45 46

o

16

41

44

00

5

6 6 6 6 7 7 7 7 9 9 9 9 10 10

0

0

1.5

38 38 38

RESULTS (5

did

Mouse not

numbers

receive

the

in parentheses initial

injection

signify

control-group

of tritiated

thymidine.

column is the number of days since the test tritiated thymidine injection. All mice received tiated (nonradioactive) thymidine the next day.

group injections

mice The

who third

received

the

with

pro-

The labeled nuclear

were

control

slides

thymidine

were itself

morphology.

observed

used

to verify

produced No

between

Downloaded from jhc.sagepub.com at UNIV OF SASKATCHEWAN on March 18, 2015

significant

cell

that

the

no significant populations

morphologic

randomly

injection alterations differences

selected

of in

198

ET.

TEICH,

from

control

slides

from test slides. As indicated peripheral clear

in

area

the

form

has

complexity is substantial.

for

our

five

to

plot

the

of age.

In

Figure

these histograms deviation ofthe

the

“central that

algorithm

used

with mean,

see

using good

derived

the vertical bars indicating that is (aI”IN , N = number

then

high 2.

this

tabulated. peak

standard P2A

This

in the

of

observed

outer

boundary-as

the

increases.

increases

slowly

has

previously

been robust

as

the

3. Bending clearly peaks

as a of

nuclear

trend

the

4.

Area

nucleus

area

of the

in

studied invariant

goes

to segmented

however, outer after

fact

is relatively

5. Perimeter after .0

nucleus-while area of the an artifact

here. This nuclear ring

Sher

opening

is consistent mass.

of the outer and then

SE, Young

starts

this

boundary-this the ring

a

to

open

and

then

IT, Zack

there

is not

a particu-

opens

parameter and then

is an

constant with

across the

boundary-this holds fairly AG, Greenberg

apparent

slight

maturation this is mature lobes. The the

range

concept

MS:

of ages

of a relatively

feature constant.

also

In preparation.

DNA

to these

features.

are from

synthesis.

two

1.5 days data set

A reasonable

populations

(P

0.15,