0022-1554191/$3.30

The Journal of Histocheznistry and Cytochemistry Copyright © 1991 by The Histochemical Society,

Vol.

39,

No.

5, pp. 607-615, 1991 Printed in USA.

Inc.

Article

Original

Uptake and Subcellular Distribution of Escherichia coli Lipopolysaccharide by Isolated Rat Type II Pneumocytes’ CRISTINA

RISCO,

Department and

JOS#{201} L. CARRASCOSA,

ofBiochemistry

Electron

andMolecular

Microscopy

Biology,

Department,

Received

forpublicationjanuary

Treatment

of isolated

Centro

25,

and

Faculty

de Biologla

1990 and

in revised

ofChemistry, Molecular,

form)uly

31,

rat Type II pneumocytes with Esche(LPS) induces a number of ultrastructural changes which become evident after 60 miii of incubation. By using post-embedding immunolabeling methods and electron microscopy, we have followed the fate of LPS after different times of incubation. After an initial period of accumulation in the pneumocyte miaovilhi, the II’S molecules enter the cytoplasm, forming discrete patches which are dispersed in some areas. After longer incubation times, LPS localize in condensed chromatin-free areas inside the nudei. LPS micelles were visualized after freezefracture and compared with the LPS-labeled membrane areas, showing that LPS micelles aggregate in particular membrane

richia coli lipopolysaccharide

Introduction

endotoxemia

(LPS), endotoxins to be involved in the and

septic

spectrum

tie is known ares

cell

formed

about on surface

binding,

distribution

after

I-ffiR or gold conjugates transport

ofLPS

and

are still scarce The lung is one

pulmonary accumulation

verse

crovascular (51)

bacterial

in mammals

components

mechanisms

exhibit

(16,36), by which

but litLPS initi-

incorporation cellular

Previous

uptake

(42) or using (HRP)-LPS

into

membranes,

and

ofLPS

have

per-

either

radioactive

been

LPS

(37), LPS antibody-

(15,25,28,31,38,49). However, studies on cell the detection ofputative cell surface binding

(25,31). of the target

organs

of endotoxin

actions.

Ad-

effects, including neutrophil, platelet, and fibrin in capillaries, endothelial disruption, increased mi-

permeability

are often

shock.

These effects

the molecular

by immunofluorescence horseradish peroxidase

(3,19,34),

sites

shock.

ofbiological

from gram-negative bactepathogenesis or induction of

damage.

Studies subccllular

A. BOSCH2

Universidad Universidad

1990

associated

(52), with

in vivo studies

and

alveolar

respiratory have shown

surfactant failure that

disruption

in LPS-induced

circulating

and November

1

Supported

UIC), 2

correspondence

should

be addressed.

accepted

1990;

Madrid,

Madrid,

Spain

Spain

(CR,MAB),

(JLC).

December

14, 1990

(0A1889).

rated

accumulate

into

Type

in alveolar

I and

Type

macrophages

II alveolar

Type II pneumonocytcs are the cells and secretion of the pulmonary surfactant 27),

whereas

change

Type

(44).

mocytes

showed

membrane

I pneumocytcs

In vitro

studies

that

phospholipid

and

biosynthesis

in Type

II cells obtained

explain

the occurrence

In this methods, treated

work the

LPS

in vitro

with

uptake, the

failure

Escherichia

surfactant

functional

during

in isolated

in plasma

and

animals

changes

could septic

by post-embedding

distribution

in gas cxType II pneu-

alterations

Ca2

LPS-trcated

ofrespiratory

we studied,

isolated

significant

(1). Moreover, from

(14).

implicated

with

fluidity,

are incorpo-

cells

involved in the synthesis in alveolar regions (12,

are mainly

performed

LPS induce

permeability

and

epithelial

partially shock

(5).

immunolabeling Type

co/i endotoxin.

II pneumocytes

Other

electron

mi-

croscopic techniques, such as freeze-fracture, negative and positive staining, lectin-gold pre-embedding, and double immunolabeling methods were also employed to study in more detail the characteristics quent

of the endotoxin-membranc intracellular

interaction

and

its subse-

distribution.

bacterial

by research grants PR84-0506-C02-02 (CR,MAB) and (JLC)from CAICYT, Spain and by the Fundaci#{243}n Ram#{243}n Areces

Spain. To whom

29,

28040

28049

zones. The sugar-specific staining in microvilli areas, where Madurapomifem agglutinin (MPA)-gold partides bind, indicates the presence of galactose derivatives in these membrane structures. Pre-treatment of pneumocytes with LPS inhibited the MPA-gold labeling, suggesting a relation between the MPA receptor and a possible LPS receptor. Finally, double immunolabeling experiments indicated an apparent LPS-tubulin association in some particular membrane regions, which could not be observed when LPS and actin were cu-localized. (J Histochcm Cytochem 39:607-615, 1991) KEY WORDS: Lipopolysaccharide; Immunogold labeling; Maclura pomifera agglutinin; Rat Type II pneumocytes.

Materials PB87-0365

Complutense, Aut#{243}noma,

endotoxins

Lipopolysaccharides na, are known a broad

MARIA

and

Methods

Reagents. Elastase was obtained from Boehringer Mannheim heim, FRG)and Percoll from Pharmacia Fine Chemicals (Uppsala, Lipopolysaccharide from Escherichia co/i 011 1:B4 was supplied (Detroit, MI). A260/A24O O.5 determined for purity verification

(MannSweden). by Difco (41).

607

Downloaded from jhc.sagepub.com at UCSF LIBRARY & CKM on March 14, 2015

608

RISCO,

An anti-LPS

in rabbits after injecting Escherichia kindly supplied by Dr. F. DIaz-Espada(ClInica Puerta de Hierro, Madrid). The antiserum specificity was verifled by a double-diffusion Ouchterlony test. Rabbit anti-tubulin antiserum was kindly supplied by Dr.J.M. Andreu (Centro de Investigaciones Biol#{243}gicas CSIC, Madrid), and rabbit anti-actin antiserum was obtained from ICN Immunobiologicals (Costa Mesa, CA). Epon 812 and Lowicryl K4M resins were from Fluka Chemie (Buchs, Switzerland) and Chemische Worke Lowi (Waldkraiburg, FRG), respectively. A 5-nm protein A-gold complex was fromJanssen (Beerse, Belgium) and 10-nm goat anti-rabbit-IgG-gold conjugate was from Sigma Chemical Co. (St. Louis, MO). A 12-nm Mac/urapom:fera agglutinin-gold complex was supplied by Polysciences (Warrington, PA). Analytical grade reagents were from Merck (Darmstadt, FRG) or Sigma. The sample sections were mounted on nickel grids covered with a layer of collodion and carbon, except for the double labeling experiments, in which gold-uncovered grids were used. All the studies were carried out using a JEOL 100 B electron microscope.

co/i.

Purified

antiserum

polyclonal

was raised

antibodies

were

Animals and Cell Preparation. Male Wistar rats weighing 250-300 g were used in all the experiments. The experiments described were performed in adherence to the CEE (86/609) and Ministerio de Agricultura (Spain) (BOE 223/1988) guidelines for care and use of laboratory animals. Type II pneumocytes were isolated from lungs of rats injected intraperitoneally with pentobarbital (10 mg/100 g body weight), by a method that includes elastase digestion and density centrifugation in a continuous gradient

ofPercoll,

pended NaCI,

as previously

in 2.6 mM sodium 5.3 mM

divided

KCI,

into aliquots,

described

phosphate

10 mM

Hepes,

(1). The

buffer, 1.9 mM

purified

cells were

pH 7.4, containing CaCl2,

and

1.3 mM

sus-

136 mM MgSO4,

and treated

with 125 at 3TC fordifferent time periods LPS was previously dispersed by sonication concentration of LPS in the buffer solution was 0.5

(15, 25, 40, 60, 90, or 120 mm).

for 1 mm.

The final

mg/mi.

incubated

without

LPS)with

the anti-LPS

CARRASCOSA,

antiserum

before

BOSCH

the gold

probe

treatment. For determination of glycocalyx integrity and the sugar residue distribution in the membranes ofType II pneumocytes, Lowicryl sections of control samples were stained with a sugar-specific stain (40) consisting of sodium phosphotungstate in an acidic medium. The distribution of Maclura pom:fera agglutinin (MPA) binding sites (galactosyl derivatives) in freshly isolated pneumocytcs was studied using MPA labeled with gold particles of 12 nm diameter in pre-embedding experiments. The pneumocyte pellets were fixed for 15 rain at room temperature with a solution ofo.5 % glutaraldehyde in PBS, washed twice with the saline buffer, and treated for another 10 mm with 0.2 M ammonium chloride. After washing with PBS, the sampies were incubated for 1 hr at room temperature either with the diluted lectin-gold

solution

complex

pre-incubated

galactopyranoside). 2%

glutaraldehyde,

ture,

and

(1:10

in PBS)

or with

1% formaldehyde

post-fixed

a solution

of the

with an excess of the hapten After washing with PBS, the samples with

1% 0504

in PBS for 30 mm

for

1 hr at room

at 4C.

lectin-gold

sugar (methylwere fixed with tempera-

Thereafter,

the sam-

pIes were dehydrated in ethanol and embedded in Epon 812. The sections were stained with uranyl acetate and observed by electron microscopy. To study the MPA-gold binding after LPS incubation, the isolated pneumocytes were incubated with LPS for 25 mm at 37C, fixed with 0.5% glutaraldehyde for 15 mm at room temperature, and processed for MPA-gold pre-embedding labeling as described above. Other samples were treated for 25 mm at 37C with LPS solutions that had been pre-incubated for 30 mm with an excess ofdifferent sugars (methyl galactopyranoside, acetyl-galactosamine, methyl-mannopyranoside, and sialic

acid).

The

incubation

was stopped

by centrifugation

and

the

pellets

fixed and processed for Lowicryl embedding and LPS immunodetection as previously indicated. For simultaneous localization oftubulin LPS and actin LPS, double munolabeling

experiments

were

done

in Lowicryl

sections

mounted

im-

on gold

Fixation and Embedding. Control and LPS-treated samples were fixed for ultrastructural studies with 2% (v/v) glutaraldehyde, 2% (w/v) tannic acid in PBS, for 1 hr at room temperature, and post-fixed 30 mm with

grids without collodion film. Each antigen was localized in one ofthe two faces of the section using gold probes of two different sizes (5 and 10 nm). After the first antigen immunodetection, the treated face was covered with a collodion film to prevent possible contamination during the second step of the experiment. After immunolabeling of the second antigen and uranyl acetate staining, the samples were observed by electron microscopy. Tubulin was detected with an anti-tubulin antiserum and 10-nm antiIgG-gold. Actin was spotted by using an anti-actin antiserum before incu-

1% 0504

bation

The incubations were stopped by centrifugation (500 rpm, 1 mm) in a table-top centrifuge and, after washing with PBS (phosphate buffer 1 mM, pH 7.2, containing 0.15 M NaCI, 2.5 mM KC1, 2 mM CaCl2, and 1 mM MgCl2), the pelleted cells were processed for electron microscopy.

at 4’C.

After

sequential

dehydration

in ethanol

the samples

were

embedded in Epon 812, polymerized at 65’C for 2 days, and sectioned. After staining with uranyl acetate and lead citrate, the samples were studied

by electron

microscopy.

Immunocytochemical

studies

for low-temperature embedding with 0.5% glutaraldehyde, 2%

were made

after processing

the cell pellets

(9). After fixation under mild conditions tannic acid in PBS for 10 mm and further

bumin

solution

in PBS.

After

jet-washing

with

PBS,

ofLPS,

the sec-

20 mm and treated 1:100 in a 1% ovalthe

samples

were

of the

in-

cubated for 1 hr at room temperature in a solution of protein A-gold (parcide diameter 5 nm)or anti-IgG-gold(10 nm)diluted 1:30 in 1% ovalbumin. After jet-washing in PBS and water, staining was performed with uranyl acetate (when employing a 5-nm protein A-gold conjugate) or with uranyl acetate and lead citrate (after using the 10-nm anti-IgG-gold complex). The immunodetection specificity was confirmed using a pre-immune serum in LPS-treated samples and by incubation of control sections (cells

10-nm

anti-IgG-gold

conjugates.

The

endo-

with anti-LPS antiserum and using a 5-nm protein A-gold conjugate as described above. The immunolabeling specificity was confirmed by using a pre-immune serum. Morphometrical

Analysis. LPS micelles obtained after sonication of iisolutions (1 mg/mI sodium phosphate buffer 2.6 mM, pH 7.4, containing 136 mM NaCI, 5.3 mM KCI, 10 mM Hepes, 1.9 mM CaCl2, and 1.3 mM MgSO4) were studied by freeze-fracture and negative staining. The LPS suspensions were frozen in liquid nitrogen and traitsferred into freeze-fracture equipment (Baizers 400T unit). The platinum/carbon replicas of the fractured surfaces were studied by electron microscopy. For negative staining, the 125 micelles were attached to carbon-costed copper grids, made hydrophilic by glow discharge, and negatively stained with 2% uranyl acetate or 1% sodium phosphotungstate in water. The electron micrographs of the LPS structures obtained by these two techniques were observed in a Nikon profile projector 6CT2 and the diameter ofthe LPS micelles determined. The pneumocyte membrane areas occupied by the endotoxin were also measured in the profile projector, using the electron micrographs obtained with the LPS-immunolabeled samples to compare them with the size of the LPS micelles. More than 150 micelles or labeled areas were studied in each case. popolysaccharide

treatment with 0.2 M ammonium chloride, the samples were dehydrated in ethanol and embedded in Lowicryl K4M at - 20C. After polymerization with UV light and sectioning, the samples were mounted in nickel grids and processed for immunodetection of LPS. Immunoelectron Microscopy. For immunolocalization tions were placed on a drop ofovalbumin(1% in PBS)for overnight at 4’C with IgG-anti-LPS preparation diluted

in the presence

toxin was detected

Downloaded from jhc.sagepub.com at UCSF LIBRARY & CKM on March 14, 2015

LPS DISTRIBUTION

IN

RAT

TYPE

II PNEUMOCYTES

609

Results LPS Effects Freshly

on Type II Pneumocytes

isolated

lar elements

Type

ure la) underwent longer

than

a

____

The

LPS.

cellular

and

localization

in this

label,

(Figure

gregates

membrane

of LPS after 120 mm)

mainly

2a),

and

different

and

lb-d).

incubation

was studied

a weak

periods

by immunolabeling

as described

in Materials

label

and

ofaggregated

After

mm

40 gold

aggregates

particles

gold

The

gold

particle

in the disperse

(Figure

remained

in the

cytoplasm

which

2d), where Cytoplasmic

number

ofgold

2c), and most inside

not

(Figure

of LPS the cyto2c), although could

some

reach

the

nu-

scattered turned

cells observed

the nucleus.

(90 and 120 mm) showed defined nuclear regions. the condensed chromatin

After 60 mm of incubation process. The cells displayed

some label. of an ordered

particles labeling

ofthe

were

endotoxin entry began to be important vesicles of variable size, probably lamel-

lar bodies, also exhibited we could observe the end mitochondrial

in ag-

structure

presence

was rather

mi-

present groups

membranous

of incubation

pneumocyte

were

label

cleus (Figure at this time.

oflabel

on the

labeled

by any distinguishable

plasmic

a large

for

included

enlargement (Figures

concentrated the

all over the cytoplasm.

surrounded

The

(Figtreated

was detected on the pneumocyte cell membrane, mainly on the microvilli. At 25 mm, there was a considerable in-

crovilli

2b).

when

alterations

mitochondria

of endotoxin in Lowicryl sections, Methods. After LPS treatment for 15 mm,

crease

changes

nuclear

cellu-

mitochondria)

in Type II Pneumocytes

(15, 25, 40, 60, 90, and

particles located

characteristic

and

Ultrastructural

of abnormal

LPS Distribution

bodies,

ultrastructural

with

reticulum

appearance

showing

lamellar

drastic

60 mm

endoplasmic the

II pneumocytcs

(microvilli,

Cells

throughout

out

(Figure

a considerable

amount

contained treated

the cytoplasm.

to be remarkable with

LPS for longer

times

a clear accumulation ofendotoxin in wellNo endotoxin appeared to be located on (Figure 2f).

The time-dependent LPS distribution in Type II pneumocytes was also expressed in a quantitative form. At least 80 cells were observed,

and

with

in a defined

label

3. We could in the ofLPS siderablc labeling

.

ci

#{149}‘“.

.

a’.

results

observe

plasma groups

sion ofthe

the

obtained

subccllular a significant

in terms

of percentage

localization

arc shown

accumulation

of cells in Figure

of the endotoxin

membrane (25 mm), followed by the appearance inside the cytoplasm (25 and 40 mm) and a disper-

endotoxin

cytoplasmic

scattering ofLPS ofthe mitochondria

label

(40

mm).

Only

after

a con-

in the cytoplasm (60 mm), was notable and nucleus observed (90 and 120 mm).

Figure 1. LPS effects on type II pneumocyte ultrastructure. (a) Freshly isolated Type II cells undergo a drastic change when treated for longer than 60 mm with LPS. Some ofthe observed alternations were (b,c) endoplasmic reticulum and nuclear membrane enlargement and (d)the appearance of abnormal mitochondna. Ib, lamellar bodies; m, mitochondria; er, endoplasmic reticulum; nm, nuclear membrane. Original magnifications: a x 8000; b-d x 16,000. Bars: a = I ram; b-d = 0.5 rim.

Downloaded from jhc.sagepub.com at UCSF LIBRARY & CKM on March 14, 2015

RISCO,

610

.5

CARRASCOSA,

BOSCH

.

.1, ‘:

n

b

f._

C

...... .

.

.

..

.

_#{149}n

.....

.

.

‘:, .

.

.

.

F,

it.

.

$.

..

J

f

Figure 2. LPS distribution in Type II pneumocytes. After incubation for 25 mm, the endotoxin was mainly located (a) on the pneumocyte microvilli and (b) forming aggregates all over the cytoplasm. At 40 mm, the gold label was rather disperse and begin to appear inside the nucleus (d). Cells treated with LPS for 60 mm or longer exhibited notable labeling inside the mitochondria(e), as well as a larger number ofgold particles scattered throughout the cytoplasm or accumulated in well-defined nuclear regions(f). n, nucleus. Gold particle size 10 nm(e)and 5 nm(a-d,f). Originalmagnifications: a,b,,f x 45,000;(c,d) x 70,000. Bars = 200 nm.

(C)

% ioo

LPS-Membrane 15mm

Size ofthe Veside

50

Size.

Interaction Membrane-labeled

Comparison (after

ical or elliptical, and of heterogeneous size pies (Figure 4A) and a more homogeneous around

25

20-50

nm;

Figure

plex structures, such hand, LPS structures 1

heterogeneous:

1

With

mm

LPS

sonication)

4B) in sonicated

as ribbons, observed

filamentous

in non-sonicated samsize (mean diameter Other

com-

were not observed. On the after negative staining were

other more

ribbons,

preparations.

discs,

lattices,

and vesicle-like

structures could be seen. The latter were of a size similar to the micelles observed in the freeze-fracture images. Negative staining has been extensively used and is still widely employed in LPS struc-

.1 120

40

ture studies (26,30,43). be seen in LPS structures (7).

iLft

and

used in the cell treatment was first studied to correlate their shape and size with the labeled areas described previously. Samples studicd by freeze-fracture showed that LPS micelles were mainly spher-

L[b-

III

Areas

The size of the LPS micelles

m gcsc

v mi n localization

We suggest

that

However, changes in shape and size could after different negative staining treatments the genuine

shape

ofLPS

micelles

corresponds

Figure a Time-dependent LPS distribution in Type II pneumocytes as percentage ofcellswith label in a defined subcellularlocalization. membrane; go, groups in cytoplasm; sc, scattering in cytoplasm; mi, mitochondria; n, nucleus.

Downloaded from jhc.sagepub.com at UCSF LIBRARY & CKM on March 14, 2015

expressed m, plasma v, vesicles;

LPS DISTRIBUTION

IN RAT

TYPE

II PNEUMOCYTES

611

30

tive residues

.4

nal

was

amounts

20

in pncumocyte

very intense of alveolar

taminants 10

in the preparation,

binding. LPS pre-treatment crease retain

microvilli

(Figure

Sb). This

lectin

sig-

in Type II pneumocytes, whereas the low macrophages, which appeared as minor condisplayed

a rather

of pneumocytes

moderate

induced

MPA-gold

a significant

de-

in MPA-gold binding (Figure Sc). However, macrophages the label observed in control samples. Pre-incubation of the

endotoxin

with

different

galactosaminc, its binding

to Type

Simultaneous and

sugars,

such

methyl-mannose,

Cytoskeletal

of LPS

Elements to correlate

components,

acetyl-

did not inhibit

II pneumocytes.

Localization

In an attempt

as methylgalactose,

and sialic acid,

the

the

LPS binding

simultaneous

labeling

sites

with

cytoskeletal

of LPS and

tubulin

was

studied. We could not find evidence of a co-localization of both molecules in internal cytoplasmic areas, but some cell membrane regions exhibited an apparent association of LPS and areas showing specific labeling for tubulin (Figures 6a and 6b). This association

was not

beled

found

(Figures

when

6c and

actin

and

LPS were

simultaneously

Ia-

6d).

Discussion 1

2 3 4 5 6 7 8 9 1011

1213161516171e

The high a specific

19

diameter

(nm)

Figure 4. Size ofthe membrane-labeled areas and comparison with LPS vesicle size. LPS samples (A, without sonication; B, after 1-mm sonication) were expressed as percentage of vesicles with a defined size. The pneumocyte plasma membrane areas occupied by LPS presented a size larger than that ofthe LPS vesicles (C). The size distribution ofthe labeled areas In microvilli (D) was similar to the LPS-located regions in other membrane zones. Size classes(nm): 1, 10-20; 2, >20-30; 3, >30-40; 4, >40-50; 5, >50-60; 6, >60-70; 7, >70-80; 8, >80-90; 9, >90-100; 10, >100-110; 11, >110-150; 12, >150-200; 13, >200-250; 14, >250-300; 15, >300-350; 16, >350-400; 17, >400-450; 18, >450-500; 19:>500.

dose ofLPS employed labeling that allows

ing

observed

by freeze-fracture

may introduce The pncumocyte

larger

than

act in each the size zones

studies,

because

trastructural than

membrane

areas

the LPS micelles,

suggesting

that

location

of the (Figures

(Figure

labeled

4C).

areas

4C and

negative

stain-

There

in microvilli

occupied several

by LPS are micelles

was no difference and

in other

interbetween

membrane

4D).

alterations

occurred

LPS-Membrane nial, which

cytes type

MPA-Gold

ervation in mild

Glycocalyx

in sections is a suitable

shows

Binding.

of freshly conditions

Sugar-specific

staining

showed

a good

pres-

isolated pneumocyte glycocalyx, after fixation and embedding in Lowicryl K4M. In particu-

lar, the pneumocyte

microvilli

exhibited

consistent

staining

(Fig-

ure 5a) which could be due to the presence ofoligosaccharide chains in the membrane structures where endotoxin is accumulated. MPA-gold

treatment

revealed

the presence

of galactose

deny-

some

that

ofccll.

oflow

method

cells were

treated

for longer

One

temperature-embedded

for antigen

aspects

be involved

ofthese

points

in the LPS-cell

concerning

ofinterest

in biological

of LPS in treated in Type II pncumothe

in the endotoxin

membrane

mate-

detection

to study the distribution distribution ofLPS

interesting

might

may take part

refers

cellular

action

con-

on this

to the factors

interaction.

that

Freeze-fracture

analysis of the LPS micelles and measurement of areas occupied by LPS in the pncumocyte cell membrane showed that LPS micelles bind

Integrity

when

Interaction

Immunolabeling

and

brane. The

Pneumocyte

to obtain At short

60 minutes.

stituents

artifacts. plasma

was required conclusions.

times ofexposure the isolated Type II pneumocytes incubated with LPS did not show any sign of cytotoxic lesions, although large ul-

samples (8), was used cells. The time-dependent to that

in this work unambiguous

aggregate preference

membranes.

and

the

pneumocytes

of a particular

The sugar-specific

the accumulation viously defined inhibition

of the pneumocytc microvilli,

mem-

where

can be clearly seen mainly after 25 mm has been reported as the time ofhighest

to isolated

to the presence

areas

of LPS for pncumocytc

toxin accumulates, incubation, which LPS binding

in well-defined

sugar

(1). This moiety

stain in these membrane

Downloaded from jhc.sagepub.com at UCSF LIBRARY & CKM on March 14, 2015

binding

be related

in the pneumocyte

at this point ofMac/urapom:fera as a specific Type II pneumocyte of MPA-gold

fact could

to pneumocytes

cell

structures,

agglutinin,

marker

endoof LPS [‘4C]-

pre(13,33), by LPS

RISCO,

612

CARRASCOSA,

BOSCH

Figure 5. Pneumocyte glycocalyx integrity and MPA-gold binding. Sugar-specific staining in the pneumocyte microvilli (a) (arrow indicates the membrane region that is not stained with PTA) where Maclure pomifera agglutinin (MPA)-gold particles accumulate (b). LPS pre-treatment of pneumocytes induced asignificantdecrease in MPA-gold binding. pm; plasma membrane. Original magnification: a x 44,000; b,c x 22,500. Bars = 200 nm.

(C)

I

1

-:i

i,

.

.

.

-

a

I p

‘a;.

b

‘.:.i

. ;;.

.

:

E?#{188} pre-treatment

support

the

‘

involvement

of sugar

structures

in LPS

concentrated

binding.

in the recycling

It is generally assumed that the hydrophobic moiety (lipid A) ofLPS is the region responsible for most ofthe cell responses elicited

for endotoxin

by endotoxins (32,35), residues in LPS binding toneal

macrophages

ton interactions to lipid

but

the

of membrane demonstrated in rabbit

was previously

(18) and could

participation

human

monocytes

rely on lectin-sugar

A recognition,

and

type

a physiological

(10).

These

reactions

relevance

sugar penrecep-

unrelated

has been

a-galactose

ments,

in Type

(22,33).

Therefore,

the

sug-

presence

ofthe

to thelamellar

brane

[‘4C]-LPS

saturable, There

from

LPS inhibition

the MPA receptor

might

II pneumocytes.

This

of MPA

also be related glycoprotein,

binding

expeni-

to the LPS receptor which

appears

microvilhi,

surfactant

bodies(33).

could

going

The existence

from

be involved the cell mem-

ofa possible

receptor

and partially reversible. are many reports on the existence blood

but the demonstration II pneumocyte

and the immune ofthc

membranes

existence could

ing to tissue

target

cells (unrelated

has

been

described

usually

of specific

system

to blood

as an unspecific

LPS reccp-

(21,23,29,45,46,48),

ofa specific

be ofgneat

from

receptor

interest,

in Type

as LPS bind-

or immune process

system

cells)

(2,17,39).

of

does not seem to be the result of a general pattern in Type II pneumocyte membrane. LfLPS and to the same pneumocyte membrane glycoprotein, as

be suggested

of the

in rat Type II pneumocytes cannot be dismissed binding experiments (1) as it was time-dependent,

derivatives

glycosylation MPA bind could

membrane

areas

processes

tons in cells from

gested (18). Recent published papers report on a unique membrane glycoprotein as the Maclura pomifera agglutinin receptor in rat Type II pneumocyte

in certain

to be

LPSLocalization

Cytoplasmic Our

data

indicate that LPS is present in the cytoplasm forming and can eventually get into the nucleus, where it ac-

aggregates

cumulates bodies,

after and

longer

endoplasmic

Downloaded from jhc.sagepub.com at UCSF LIBRARY & CKM on March 14, 2015

incubation reticulum

times. are the

Mitochondnia, distinguishable

lamellar struc-

LPS DISTRIBUTION

IN

II PNEUMOCYTES

TYPE

RAT

rT

Figure6. Simultaneouslocalization of LPS and cytoskeletal elements. Double immunolabeling experiments showed an apparent association ofLPSwlthtubulin in some cell membrane regions (arrows in a and b), which did not occur when actin and LPS were simultaneously labeled (c,d). Protein A-gold(5 nm)wasemployedto label the endotoxin, whereas 10-nm gold particles reveal the tubulin or actin localization. Original magnification x 60,000. Bar = 200 nm.

613

...::...:.

.

..

#{149}

.‘.j.

e.

#{149}

.

.

..

::;s.

:a

.

.

.

I..

.

4:

.&

V.

.

-

!

.

...

-.;#{149}

S..

..

a#{149} ,

.

.

.

.

.

.

:

.

:“#{149} C

tunes

in which

alterations viously

LPS can be observed

ofmitochondnial described

in detail

(6), and

cellular interactions ofLPS, cell membrane disruption The transport ofendotoxin be related to intracellular problems associated

despite

could

be related

the possible

(4,24). aggregates

across

to direct

the presence

ofcell

co-localization

experiments)

membrane

(never

suggests

observed

the possible

teraction we have

in endotoxin-tneatcd observed that brain

strongly

inhibited

interaction

network, tion could in endotoxic

occurs

regions

existence

explain many processes.

of the

could

defined by a complex network and still poorly participate in this LPS preference for some

but techthin un-

to confirm the experiments. double

ofan

complex

amounts than cleus and alter by which pression

We have

remain

tan

Type

that

observed

ofdefincd

aggregates

cumulation

ofthis

II pneumocyte

Further

LPS is selectively

nuclear

regions.

accumulated The

presence

in cerof the en-

in physio-

(in much

synthesis

membrane

regions

lower

Finally,

the possible

on the detailed

charide

in Type

Downloaded from jhc.sagepub.com at UCSF LIBRARY & CKM on March 14, 2015

gene

cx-

aspects

mechanisms

II pncumocytes

by which as well

at which

into the cytoplasm

areas situated near model of endotoxin

the primary

and

some possible ways in Type II pneumocytes,

component in particular the MPA binding glycoprotein and a putative “LPS receptor”

us to understand toxicity

be significant

bacterial

our results.

studies

plasma

and translocates

A relation between mocyte membranes from

(50), could regions.

the endotoxin

in DNA

as the existence

dotoxin

described nuclear

(20,47).

be

gested

the

alterations unknown

In conclusion, we have defined LPS can elicit its direct toxic effects

the Cell Nucleus observed

in which

in-

the doses used in our studies) could get into the nuthe nuclear function, but the molecular mechanisms

LPS induces still

of LPS could

processes,

in-

Ifthis strong microtubular

events

action

endotoxic

LPS-tu-

of this disorganiza-

cellular

intranuclear

logical

labeling

LPS-tubulin

data). ofthe

The

cells (21). The existence of alterations in ionic distribution or the intranuclear compartmentalization,

iion in cytoplasmic suggests an attractive

LPS and

(25,31) and the existence been suggested in tumor

the cytoplasm

ofapparent

effects

has

macrophage tranuclear

Type II pneumocytes. Furthermore, microtubulc polymerization can

intracellular

has been reported

H1 interaction

from

in vitro by LPS (unpublished in vivo and causes a disruption

the subsequent

in the nucleus

of an LPS-histone

effects derived

in LPS-aCtin

intra-

.1I

...

dotoxin

the cell. Functional

systems have been pre-

network in sections (1 1) make it difficult ofLPS-tubulin association in doublelabeling

However, bulin

inside

structures such as microtubules, with the localization of this

nical

oriented existence

once

and microsomal

-

#{149}

..

‘,

$

.

1’#{149}#{149} .1’

,..

nudear regions. ofType II pneucan also be sug-

LPS-tubulin the plasma action. outlined of bacterial and

other

en-

and the ac-

associamembrane

above

can help

lipopolysactarget

cells.

614

RISCO,

Cited

literature 1. Aracil

FM,

Bosch

MA,

Municio

AM:

Influence

of E. coli lipopolysac-

charide binding to nat alveolar type II cells on their functional ties. Mol Cell Biochem 68:59, 1985

proper-

3. Bona CA: Fate ofendotoxin in macrophages: tural aspects. J Infrct Dis 128:74, 1973

biological

20. Halleck MS, Schlegel RA, Rose KM: Cytoplasmic to nuclear translocaiion ofRNA polymenase I is required for lipopolysaccharide-induced nucleolar RNA synthesis and subsequent DNA synthesis in munine B lymphocytes. ) Cell Sci 92:101, 1989 21.

2. BaraJ, Lailler R, Trudel M, Brailovsky C, Nigam VN: Molecular models on the insertion of a Salmonella minnesota form glycolipid into the cell membrane ofnormal and transformed cells. EurJ Biochem 35:495, 1973 and ultrastruc-

4. Bosch MA, GarcIa R, Pagani R, Portol#{233}s MT, DIaz-Laviada I, Abarca 5, Ainaga MJ, Risco C, Municio AM: Induction ofreversible shock by E. coli lipopolysaccharide in rats. Changes in serum and cell membrane parameters. Br J Exp Pathol 69:805, 1988

SG: Cellular and molecular mechanism Annu Rev Microbiol 33:67, 1979

of action

of bacterial

endotoxins. 7. Brodgen

KA, Phillips M: The ultrastructunal morphology of endotoxins and lipopolysaccharides. Electron Microsc Rev 1:261, 1988

8. CarrascosaJL: tron Microsc

Immunoclectron Rev 1:1, 1988

microscopical

studies

on viruses.

Elec-

9. CarrascosaJL, Gonzalez P, Carrascosa AL, GarcIa-Barreno B, Enjuanes L, Vifluela E: Localization of structural proteins in African swine fever virus particles by immunoelectron microscopy. J Virol 58:377, 1986 10. CavaillonJM, Fitting C, Hauttecour B, Haeffner-Cavaillon N: Inhibition by gangliosides ofthe specific binding oflipopolysaccharide (LPS) to human monocytes prevents LPS-induced intenleukin-1 production. Cell Immunol 106:293, 1987 11. Dc Mey J, Langanger G, Gcvens G, Nuydens R, Dc Brabander M: Preembedding for localization by electron microscopy of cytoskeletal antigens

in cultured

Methods

Enzymol

cell monolayers 134:592, 1986

12. Dobbs LG, Mason RJ, Williams of surfactant by primary cultures rats. 13.

Biochim

Dobbs teristics

gold-labeled

MC, Benson B), Sveiski K: Secretion of alveolar type II cells isolated from

713:118,

Biophys

Acta 846:155,

during

shock.

Virchows

Arch

in biochemical charactype II cells with time

1985

404:197,

CH, Bancells in the

1984

FuerstJA, Perry)W: Demonstration oflipopolysaccharide on sheathed flagella ofVibnio cholerae 0:1 by protein A-gold immunoelectron microscopy. J Bacteniol 170:1488, 1988

GarcIa

R, Vilonia

ACTH

induced

MD, adrenal

Municio AM: Influence ofE. coli endotoxin on cell steroidogenesis. J Steroid Biochem 22:377,

1985 18. Haeffner-Cavaillon Specific

rophages: 19.

binding

serum

Haeffner-Cavaillon chanide receptor

tenistics.

N, CavaillonJM, Etievant M, Lebbar 5, Szabo L: of endotoxin to human monocytes and mouse macrequirement. Cell Immunol 91:119, 1985 N, Chaby

R, CavaillonJM,

on rabbit peritoneal J Immunol 128:1950, 1982

Golenbock

Szabo

macrophages.

L: Lipopolysac-

I. Binding

charac-

DT,

of macrophage

23.

Raetz

tumor

Kabir

5, Rosenstreich

DL:

spleen

lymphocytes.

Infect

CRH:

Lipid

cells. J Biol Chem

A binding 263:14802,

sites in 1988

Binding

Immunol

of bacterial 15:156,

endotoxin 1977

to munine

24. Kabir 5, Rosenstreich DL, Mergenhagen SE: Bacterial endotoxins and cell membranes. In Jeljaszewicz J, Wadstr#{246}mT, eds. Bacterial toxins and cell membranes. New York, Academic Press, 1978, 59 25.

Kang YH, Carl M, Maheshwani RK, Watson 12, Yaffe L, Gnimley PM: Incorporation of bacterial lipopolysaccharide by human Leu-11a natural killer cells: ultrastructural and functional correlations. Lab Invest 58:196,

1988

26. Kato N, Ohta M, Kido N, Ito H, Naito 5, Kuno ‘P Formation of a hexagonal lattice structure by an R-form lipopoiysaccharide of KIebsiella sp. J Bacteniol 162:1142, 1985 27.

Kikkawa

Y, Smith F: Biology ofdisease.

ofpulmonary

surfactant

in health

Cellular

and disease.

and biochemical

aspects

Lab Invest 49:122,

1983

28. Lam JS, Lam MYC, McDonald LA, Hancock REW: Visualization Pseudomonas aeruginosa 0 antigens by using a protein A-dextran-colloidal gold conjugate. J Bacteniol 169:3531, 1987 29.

of

Lei MG, Morrison DC: Specific endotoxic lipopolysaccharide-binding proteins on munine splenocytes. I. Detection of lipopolysaccharidebinding sites on splenocytes and splenocytes subpopulations. Immunol 141:996,

1988

30. LopesJ, Inniss WE: Electron microscopic study of lipopolysacchanide from an avian strain ofEschenichia coli 018. J Bacteniol 103:238, 1970 31. Lucas RM, Subramoniam A, AleoJJ: Intracellular localization of bactenial lipopolysaccharide using the avidin biotin method at the dcctron microscopic level. J Peniodontol 56:553, 1985 32.

L#{252}denitz0, Freudenberg

Shaw DH: Microbial

MA,

membrane

rent topics in membranes 1982, 17:79

16. Galanos C, Freudenberg MA, Matsura M, Coumbos A: Hypersensitivity to endotoxin and mechanisms ofhost response. In LevinJ, Ten Cate Jw, Stunk A, Buller HR, eds. Bacterial endotoxins: pathophysiological effects, clinical significance, and pharmacological control. New York, Alan R Liss, 1988, 295 17.

RY,

22. Joyce-Brady MF, Marshall B, BrodyJS: Pulmonary alveolar type 2 cells from three species exhibit a predominant high molecular weight MPA binding glycoprotein. J Cell Biol 101:104a, 1985

1982

MC, Branot AE: Changes oflectin binding ofalveolar

Biochim

antibodies.

Freudenberg N, Freudenberg MA, Guzman), Mittermayer dara K, Galanos C: Identification of endotoxin positive rat lung

15.

Acta

LG, Williams and patterns

in culture. 14.

Biophys

using

Hampton

membrane

5. Bosch MA, Risco C, Municio AM: Effect of Escherichia coli lipopolysaccharide on phosphatidylcholine biosynthesis by rat lung and alveolar type II cells. Mol Cell Biochem 93:167, 1990 6. Bradley

BOSCH

CARRASCOSA,

33. Marshall BC, Joyce-Brady ization ofthe pulmonary tinin-binding

membrane

and

C, Lchmann V, Rietschel ET, In Razin 5, Rottem 5, eds. Curtransport. New York, Academic Press, Galanos

lipids.

MF, BrodyJS: Identification alveolar type II cell Maclura glycoprotein.

Biochim

and characterpomifera agglu-

Biophys

Acta

966:403,

1988 34. MathisonJC, Ulevitch RJ: The clearance, lar localization ofintravenously injected J Immunol 123:2133, 1979

tissue distribution lipopolysaccharide

and celluin rabbits.

35. Matsuwra M, Kozima Y, HommaJ, Kubota Y, Yamamoto A, Kiso M, Hasegawa A: Biological activities ofchemically synthesized analogues ofthe non-reducing sugar moiety oflipid A. FEBS Lett 167:226, 1984 36. Nowotny A: Review of the molecular tions. Rev Infect Dis 9:503, 1987 37.

requirements

of endotoxic

ac-

Nygren H, Daglen G: Ukrastructural localization of lipopolysacchanidebinding sites with peroxidasc-conjugated lipopolysaccharides. J Immunol Methods 25:335, 1979

38. Odeyale CO, Kang YH: Biotinylation of bacterial lipopolysaccharide and its applications to electron microscopy. J Histochem Cytochem 36:1131, 1988 39. Pagani R, Portol#{233}s MT. Municio AM: The binding of Escherichia endotoxin to isolated rat hepatocytes. FEBS Lett 131:103, 1981

coli

40. Ramboung A: Morphological and histochemical aspects of glycoproteins at the surface of animal cells. mt Rev Cytol 31:57, 1971

Downloaded from jhc.sagepub.com at UCSF LIBRARY & CKM on March 14, 2015

LPS DISTRIBUTION

41.

Romanowska

IN

RAT

E: Sepharose

lipopolysaccharides.

Anal

TYPE

gel

II PNEUMOCYTES

filtration

Biochem

33:383,

method

615

of purification

of

pression

1970

42. Rubinstein HS, Fein J, Coons AH: Localization of endotoxin walls of the peripheral vascular system during lethal endotoxemia. Soc Exp Biol Med 111:458, 1962

in the Proc

48.

43. ShandsJWJr, lated bacterial

ofiso-

49. Uragoh

GrahamJA, Nath K: The morphologic lipopolysaccharide. J Mol Biol 25:15,

in munine

by agents

structure 1967

peritoneal

that elevate

macrophages

intracellular

Tobias PS, MathisonJC, UlevitchJC: binding proteins involved in responses Chem 263:13479, 1988 K, Sueshi

histochemical

K, Nakamura

method

is selectively

CAMP. J Immunol

suppressed

142:1274,

T, Iwanaga

for in vivo detection

S: A novel

ofendotoxin

immuno-

using

44. Sorokin SP: The cells of the lung. In Nettesheim P, ed. Morphology ofexperimental respiratory carcinogenesis. AEC Symposium Series 21:3, 1970

shOd crab factor C. J Histochem Cytochem 36:1275, 1988 50. Verheinjen R, Van Venroojj W, Ramaekers F: The nuclear matrix: tune and composition. J Cell Sci 90:11, 1988

45. Springer GF, AdyeJC: kocytes and platelets.

51.

Endotoxin binding substances from human Infect Immunol 12:978, 1975

46. Springer

GF, Adye JC, Berkorovainy activity of the lipopolysacchanide-receptor Biochemistry 13:1379, 1974

47. Tannenbaum

CS, Hamilton

leu-

A, Jirgensous B: Properties and from human erythrocytes.

Von Wichert remodeling

of septic

in rat lung.

shock Biochim

upon

horse-

struc-

phosphatidyicholine

Biophys

Acta

664:487,

1981

52. Worthen GS, Haslett C, Smedly LA, Dees AJ, Gumbay S, HensonJE, Henson PM: Long vascular injury induced by chemotactic factors: enhancement

1k: Lipopolysaccharide-induced

P: Influence mechanism

1989

A family of lipopolysaccharide to gram-negative sepsis. J Biol

by bacterial

gene cx-

Downloaded from jhc.sagepub.com at UCSF LIBRARY & CKM on March 14, 2015

endotoxins.

Fed

Proc

45:7,

1986