VVrrJBViJVVVV: I

i 33... 30‘... 3! t I :4: V L3.oo.b§n.29iu:l .I... 1 1"

...... V. 3-. ...-9.... - tool!

Vuhmflflmnfiwiraf “Kata“...

J. ... I. . V V
.38... V». 8. . . V VVVIQ?2V-...!’..:C~.:t&.. :0 ......‘V
. V 30.0.10..- .V. V V... .9L.u0¢!( V V
V ...-...}: V V V. V V V. V ..V
V . V . V . V V V-.\¢.V: Vi. 0.
V . ...:3h ...-.75.. . V V ..VVtVo.V.I:.
iv... 3.... ...: . V V ....oV¢VJ~.'V.".

350‘1.o'-.Q¢¢>.h.' .{n v...
la.-."3..s.}oo. ‘0. .
... ’353..$.V-.!.I’1: o t...-
848.133.,"
wgcd::¢.o.o"‘ . V
«1-40-3430... V V . on... '3' ‘xu... .
. . . . ... loo. 3:! 1
V . V ..3~.’DV.V' ...u.
)3..;V..‘;o:’o ...". V5...- V
sooa"'J-o53'v€‘ . (no.
‘.oi..oo.-
zilzzl... 1.2.51.1! . V V a. V I!
‘ “O V V ' u . n V ..ths.‘ 00., V o n . .
Coo-a3 Ono—...“. .JJV‘ I. V V V fiuVol€Oan¢IIVVoVY (-02: V V . V V V V
V V V V .39. .- V99...V1v~ .....oV!.VIV 713 V d. ‘1..va
. V . 30:0... I; V .- I ......(f V. V V V. V V V V V ‘59.... 0... anti... 0 .I-V.
. 72......) oV‘QVVSI. V V . V. . V V V V 9:10.33 . Vii-...? 0”
2r!.- 1.00.83... ... V .V 1.7%. P '63. V V?!
V .‘U..“V‘ Q. 0

oil. naI‘I... I... :IVOVQ“, ... v.
....‘f‘oo ... 0.730.... 3.133;:
.... 733......1.‘ 0‘: 30.23733.
...:83...

1 ‘09-‘00;
.. .13.... '9-

.uvosV 0. V
.....Voc

(9!. .41313‘. .‘rV. .
... rutqu-‘Iv... it: 3.4.
Vi. (‘3..31: 32c. .V... 53:!“ .
...‘lV.L-ta.o...o: 2). 0. V V . V . V. V V V. .V V V V .
o‘OIOV‘o-i. V . V V N . . V V . ..nV.
V V. . V V V . V 9.0. V‘ V
.V V V. V V . . V 9 .2' V V
lot-.5302! .53113... ...... V 9.. v .\V V V.
111.03. I..V.l.-:‘..VO‘. . V. . ...?!- 3 .0.
‘7 G... 32:11.33: . . y.

71¢..n. . V . VV V V . V I. VVSVVVI...I:
.....u .5 . V . V V V . V . V . . V V V. V V . . 'Q . V ... 0.: ...-... ..- VcV
‘0. IoV- V V . .4: V V V V V V . V V V V V V V. V V 9.. y ..OVVOVH’. "37‘.
V V .0". u ‘0 VI- . . V . V. V . . . . . V V V V V . ...I’. . 0‘, ...... 9". Vt:"vtkr:
. ......V . .V . . . . _ . . _ . . . .V V. V V. . . . .1.» ..VV. ...VVVVV .VVVVVtVVV:VVVn.VVb.V «....
V V ... ......t V V. V V. . . V V V V :1 s. V . ach‘.\ .I Q
oltLMAVVonozunruuahrt 9.5.0 V7.00..V.Q.I¢. . V V . V . . V. 2...)?! .5... HQ»...
:33...‘ 3:0? .....oJV .V ....o V V . V n
.....I'...V|..JIV V . V . V V V V V .V . . V. .V..¢V74.V&. w
3 ”a .3. V V V V V . V V V V V V V . . . V V V V ....V. V0...V’!oV-.9
V n .0 0‘; V. V V . V V V. VV V V V V V V V V. V .... V .
....4. V . 5.3.1.0643 V V.. .V V V . V . . V. V V _ V . V 7... ..." ...-J'u‘VO.‘
3" .V... :1 .. ...... IVOVova’ ;_ .
V V no.0...V1o‘iV ‘Vta 6:.
. V ‘V...

. V o. o. . V. . V V V V VV

‘.V:0Vo..V9‘V: . V V V V _ V V . V V V ’2' V‘In" V
. V . V V V . V V V ...-(V't‘V 1.93...

V V. V. V V V V V . . .V V V .VV .‘ fu‘oc.

V . .. V . V V . V (...-.03... V

V. V 0 ‘V . Q .

.. ...-... o . I

.V
....o a ....u.‘...

o..‘....

.0 V .
the... 5......VOA

'00

.0 0o! .00 .85.. V7.14 a...
. 3 .1...c._‘V>V.VoVV‘OSV
r.-Vl..»...o p . 3..

V . V 3...... 24.90....VV
V V V..-

.q. v :0 0.. u
. ,Q,’

E

.....L: 0V0 V V V .V .V .
. 5.. V . V V V ... V . «V n . V . V V V
.0. I .10... V V V V V V V V . . r. 0.. . V . .
V... V V V .‘-....’0 . . V ..ll-V—On7.
.
. vvh V V o“- . 5. V

'03:»... ..T‘. .....V V...VOV"o-X:..:f7 >V
VV..VVV V . ... .-

1.1.3.31.

V...
V. 9...... . ... ..toon.
...... ...... ..‘n.,s 0.9...- V
I. :1... .th‘... ...: V. 0...... .

...-91...... 13...... o. .0

. 10...; £vu.
99.0;
O...n.n.fi.”‘.9\la‘.
I“

.Von"...
V . u :0.

. V V L c v V a..." V . V V V
Vv‘ V V. V.. .‘.o V V V V V V.. V
. ... . . . V V ...... .31.... ...... V - V V. .v . V V V V V . V V ......V . . V .
a . V . ~ 0. V V ' V I . V ~ . c . .
V V 1. o A . :45-.. V a V V ..J ..Io\- 0...: o a V . . 0‘ . V..' ...-7‘.- ..A V .V. c... ’78. O. V... . . V .V V 0.. V V . VVVQ’QV. ....VV. V
VV ... VV . -.OV¢I9 V V .v0 01.. V I V. ..n. .q. .‘V 3‘. V . V V V ......
V V .V

...-2.9:). V
. 5.. I. ......V

v
.
0

W.
w
2..

n V .V V a. V
on 19. - VVVIVV .V V ..o.......
_ ...nl V . V
. .p . .1 ¢. . .VV . D. V V V
V uVQVQIIQOVo a... V. V V
t. V. 0.. -.0.. o ...-o.- V V .V V . . . V V V V .
V . .uut‘coit V . V . VV . . V V. V .o
. V . V .V .V V V V . V. V. V .V V .VVV .V V V V VV V V V ..I. V
V .. V . V. V .V V V . V V V. «V. v'cut. ...‘V
V V V VVVV V V. . V V..V..-oo ..q¢f- .VVVVVV...V0.V
V ...-.05.}... ...I.¢VVI 0......1. I
.‘o-. .....Vo. -. V t V
V .....u.

lVth

. a. l u A:
c . . r.... .V ... V V a... g V l¢.\.vl.!. .v IOV‘IVvfi. vvvvv
. xf. 10.45.. S?..~.\DVV

D. . u...

'WRWV“

...... ....VvV .....VT-oVu: to. 00.4-
....._.‘.‘o n‘...¢...vVa .ndV.V.|V4 I V V .55... u..... o . .
. . .V 4 v.1. . . V V .3
V V. V V. V. V V VV . V V V V . V. V ... V
V V . V ” V V. V . . .V. V. .a .‘.o..
. . v.... ...}IV
V. ....l.....¢.. .. V V V V...
. .V V . . VV V . .o I.-
. . V

1391‘

...~ c:
0..
V. ...
..9...I.O¢ . V V V V V V V V V V... . V
V... V‘: . VV. V V. V V . . V. . V. . V. V V V . . V V V V .V V V . VQOVV .
.. . V .V . . V V V V. V .. . . V .V . . V V.............
V ..VV... V - ¢
0V..s¢.oVoV . . ‘-
..V..o.9~. ...-......
. u: 1 II... .V-V

.

”-9-".an

O ....V of. . ‘V
6.. it . ...
o J ... I‘.\

.V . V
.- V. .V V V V V V V V
. V . . V. V V. . V V V. . V .
. V V. V. .. V. V. .V . V .....VuVaVo.
V V. V .V V. . ......‘1s; .. 05.0.. h
V V. z u .VV

V . V
...-... V 0.9Vr' - 50"
.

W

V . A.
a V .V V V
y In»... .VuV V . . V V V 7v...|l-o0n.. V
VV. V V V Va... -. .0 on. V V V V V . V V V V V V
V V V . V V . V V . V V V V . . V ~ 7.0.4:}... .
. V V ..V

w

0.0 .
V .
vo.V.o- .ch
...... . V
n

ram

.. V. V.
soft-V... . .
til-o .. a

1m

.0
I . .49. V. . .
. VVY. V V V V V . V V V
... . V V V V . V.VV . V. . V V V . . V . V .
. V . V V V V V _ . T316; .

V 2 V .V V
. V . o..V..o. ...
... V—VV"
IVQQVOI‘QI.I~0
otb.t......v.%i.914.ov3vV

....(VV..X. V .V V V .V
. V V V V ... V V. . V . V V V . V V V V .a-V
. V V V V . VV vVVVIV 0..
V ... uVV
:.vv-'§.'. 3.0.0.!an

...-lo: 0 V

.‘o. 1 I V. . ...

o...»o.o.v .V. . V V V V VV .V V V V V V
V ... .V . V V... I... V .. V V... V. ...V . . . . V ... . ... V V V V V V. V
.V .V V V V V V..s .0...

.....V.
n ...-i
V.....V.
V ... ... .
V...........
... ... .su.

..Aq...s...
V. O.... V.”
t.....—..

o

'“nvmmum

.c.
V ...
o. . 0.9.
...‘V ..t.oo’I
.0'... .
... ...
V: V. V V. .....A.VuV
.....‘tO... V V V V V c . ....
... V

«Mm ,

E

0.. V V V V V V. V
. . V . V V .V . . V. . V. V V V . V. V V . V VVvo .
. V V . V VV V . V .....o.
.. V. .I I V . V. .- - ...

PM!

V

1‘.

. . V
. ..IV‘OV ....
V V V V n. . .
a. ... n o .
V. V V

.l ..
VJ... ...? 2..
...-out...
V n. 23. V .V V V. .... . V .
. . V _ . V V. V . V. V V. V .V V V V V .20.... . .V V.
. V V VV V .V VV . . ...-'V . V .
V ...-oQVOOV- . V . V . . I. .Vol II V-
ooV~Ilnonuuq V y . ...-"-‘O
V ..V ....V.. V. V V
V...

‘

f“

v
0
o.

 

uj'fi!

V. ....V
n.

.V V .
AV . V V V V V
.... V V V. . VV . V V. V . V .
V . V V . . V .V V. .V V . V .01
V . V . V . V V0.1...‘GufionViVO
V V...

This is to certify that the
dissertation entitled

ROLE OF THE SPHINGOMYELIN/CERAMIDE PATHWAY IN
DIABETIC RETINOPATHY

presented by
MADALINA OPREANU

has been accepted towards fulfillment g
of the requirements for the j

Doctoral degree in Microbiology and Molecular
Genetics

Major Professor’s Signature
// / Q17 /2+ 2 C7 /0

Date

 

 

 

 

MSU is an Affirmative Action/Equal Opportunity Employer

 

LIBRARY
Michigan State
University

 

 

 

 

PLACE IN RETURN BOX to remove this Checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProlecc&Pres/ClRC/DateDue.indd

ROLE OF THE SPHINGOMYELIN/CERAMIDE PATHWAY IN DIABETIC
RETINOPATHY
By

Madalina Opreanu

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Microbiology and Molecular Genetics

2010

ABSTRACT

ROLE OF THE SPHINGOMYELIN/CERAMIDE PATHWAY IN DIABETIC
RETINOPATHY

By

Madalina Opreanu

I

Diabetic retinopathy (DR) stands as a sight threatening disease without effective
therapeutic options. Early DR is recognized to be a persistent low-grade chronic
inflammatory disease. Acid (ASMase) and neutral (NSMase) sphingomyelinases
(SMases) are important early responders in inflammatory cytokine signaling and key
regulatory enzymes of sphingolipid metabolism. Sphingolipids are a major component of
plasma membrane microdomains and sphingomyelin hydrolysis by SMases to bioactive
lipid ceramide represents a prerequisite for inflammatory cytokines signaling. This study
addresses the role of SMases in retinal vascular endothelium inflammation and
development of microangiopathic lesions in the retinas affected by retinopathy. We first
demonstrated that inhibition/gene silencing of both SMases decrease cytokine-induced
cellular adhesion molecules expression in human retinal endothelial cells (HREC); yet a
more pronounced anti-inflammatory effect was observed when ASMase inhibition/gene
silencing was attained. Similarly, docosahexaenoic acid (DHA), the major (03
polyunsaturated fatty acid in the retina and well known anti-inflammatory agent,
downregulated cytokine-induced inflammation in HREC. Interestingly, DHA decreased
basal and cytokine-induced ASMase and NSMase expression and activity in HREC,
further underlining the role of SMases as mediators of the inflammatory process. SMases

pathway rather than ceramide de novo synthesis pathway was important for inflammatory

signaling in HREC. To further demonstrate the role of SMase pathway in inflammation,
the caveolae/lipid microdomains sphingolipid composition was then characterized. In
support of SMases inhibition and displacement of ASMase from caveolae microdomains
by DHA, we found a significant decrease in ceramide/sphingomyelin (Cer/SM) ratio in
the caveolae fraction isolated from HREC treated with DHA; moreover, DHA prevented
the TNFa-induced increase in the Cer/SM ratio in caveolar membrane microdomains and
intracellular inflammatory signaling. To address the role of SMases in retinal
inflammation and vessel loss in DR, an in vivo model of streptozotocin (STZ)-induced
diabetes in rat was employed. ASMase, but not NSMase was upregulated in the diabetic
retinas with an inflammatory status and microangiopathy lesions. DHA-enriched diet
restored ASMase in diabetic retina to the control levels and prevented retinal
inflammation and vessel loss. More importantly, DHA supplementation specifically
prevented vascular ASMase upregulation in the retinas of a rat model with
vasodegenerative phase of DR. To directly investigate the role of ASMase in
development of retinal acellular capillaries and ocular neovascularization, we next used
accelerated models of retinopathy by inducing ischemia/ reperfusion (I/R) injury and
oxygen-induced retinopathy (OIR) in wild type (ASMase+/+) and ASMaseT mouse
retina. ASMase'f mouse retina was significantly protected against retinal inflammation,
vascular degeneration and ocular neovascularization. In summary, this is the first study to
show that ASMase, a key regulatory enzyme of sphingolipid metabolism, is a novel and
fundamental mediator and a promising therapeutic target for the prevention of retinal

vascular inflammation and diabetic retinopathy.

ACKNOWLEDGEMENTS

I would like to deeply thank my committee members Dr. Julia V. Busik, Dr.
Walter J Esselman, Dr. Andrea Amalfitano, Dr. Michele F luck, Dr. Karl L. Olson and Dr.
Richard C. Schwartz for their valuable guidance and support for all these years.
Especially I would like to dedicate my special gatitude to my mentor Dr. Julia V. Busik
who guided my steps from the very early days of my research journey, teching me with
enthusiasm and passion special techniques, opening my "mind eyes" towards a critical
thinking; nevertheless, I would like to express my deep appreciation for her frendship,
insight and understanding. I also want to express my profound gratitude for my mentor
Dr. Walter J. Esselman who always offered me with the most kind and warmest support
and encouragement so I can manage easier the uphills and downhills of my graduate
career.

Many thanks also go to Dr. Karl L. Olson and Dr. Narayanan Paraeswaran for
their unconditional opening of their labs for me to apply any resources and for their
insightful suggestions on experimental and technical apporaches. And more, Dr. Gloria
Ines Perez in the Department of Physiology who contributes tremendously to my research
by providing me the tools necessary for my experiments and who's door was always
opened for help, advice and technical support.

I am also very thanka for the fiiendship and valuable help of all my lab mates,
Svetlana Bozack, Todd A. Lydic, Maria Tickhonenko, Kelly M. McSorley, Andrew
Sochacki, Matt Faber in Dr. Busik lab. Particularly I thank Svetlana Bozack for her

unlimited friendship and technical support; and Todd Lydic for his great assistance with

iv

mass spectrometry analysis. I would also like to thank my other lab fellows from Dr.
Olson Lab and Dr. Parameswaran Lab, especially Chris Green and Sonika Patial for their
valuable discussions and friendship.

At last, but not least, I express my deep appreciation for my husband, for my
parents and my entire family for their unflagging love and tremendous support during my

graduate career.

TABLE OF CONTENTS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LIST OF TABLES x
LIST OF FIGURES , xi
LIST OF ABBREVIATIONS xiv
1. LITERATURE REVIEW 1
1. Diabetic Retinopathy 1
1.1. Overview ............................................................................................................... l
1.2. Histopathological and clinical manifestations ....................................................... 2
1.3. Epidemiology ........................................................................................................ 4
2. Dyslipidemia and diabetic retinopathy 6
2.1. Insulin and lipid metabolism ................................................................................. 6
2.2. Diabetic dyslipidemia ............................................................................................ 7

2.3. Retinal fatty acid profile and diabetic retinopathy .............................................. 11
3. Inflammatory mechanisms in diabetic retinopathy 17
3.1. Adhesion molecules in diabetic retinopathy ........................................................ 17
3.2. Proinflammatory cytokines in diabetic retinopathy ............................................ l8
4. Sphingomyelinase and Inflammation 23
4.1. Neutral sphingomyelinase ................................................................................... 23
4.2. Acid sphingomyelinase ....................................................................................... 25

4.3. Sphingomyelinases, plasma membrane microdomains and inflammatory signal
transduction ................................................................................................................ 30
5. Retinal vascular endothelium 37
5.1. Overview ............................................................................................................. 37
5.2. Sphingomyelinases and vascular endothelium .................................................... 38
6. Hypothesis and objective of thesis 40
7. References 44
II. DOCOSAHEXAENOIC ACID INHIBITS CYTOKINE SIGNALING BY
DOWNREGULATING SPHINGOMYELINASES IN HUMAN RETINAL
ENDOTHELIAL CELLS 65
1. Abstract 65
2. Introduction 67
3. Materials and Methods 70
3.1. Reagents and supplies .......................................................................................... 70

3.2. Cell culture .......................................................................................................... 71

3.3. Fatty acid treatment ............................................................................................. 71
3.4. Sphingomyelinase assay ...................................................................................... 72
3.5. MCD and cholesterol treatment ........................................................................... 72

3.6. Cholesterol measurement .................................................................................... 73

3.7. Western blot analysis ........................................................................................... 73
3.8. Gene silencing of ASMase and NSMase ............................................................. 74

vi

3.9. Quantitative real-time polymerase chain reaction ............................................... 74

 

 

3.10. Isolation of caveolae/lipid rafts membrane domains ......................................... 75
3.11. Isolation of plasma membrane domains ............................................................ 76
3.12. Lipid extraction ................................................................................................. 76
3.13. Mass spectrometry analysis ............................................................................... 76
3.14. Statistical Analysis ............................................................................................ 78
4. Results 79
4.1. Dose response of TNFa and IL-IB induced cellular adhesion molecules (CAMS)
expression in HREC ................................................................................................... 79
4.2. TNFa and IL-lB induce ASMase and NSMase activation in HREC .................. 79
4.3. Inhibition of cytokine-induced ASMase and NSMase activation by DHA in
HREC ......................................................................................................................... 80
4.4. Effect of ASMase and NSMase inhibition and/or gene silencing on TNFa and
IL-IB induced CAMS expression ................................................................................ 81
4.5. Inhibition of ASMase and NSMase activity by DHA ......................................... 82
4.6. Inhibition of ASMase and NSMase activity by MCD ......................................... 82
4.7. No effect of ceramide de novo synthesis inhibition on TNFa and IL-IB
inducedCAMs expression in HREC ........................................................................... 84
5. Discussion 106
6. References 112

 

III. DOCOSAHEXAENOIC ACID REGULATES CAVEOLAR ACID
SPHINGOMYELINASE EXPRESSION AND PROMOTES SPHINGOMYELIN
ENRICHEMENT OF MEMBRANE MICRODOMAINS IN HUMAN RETINAL

 

 

 

 

 

ENDOTHELIAL CELLS 119
1. Abstract 1 l9
2. Introduction 121
3. Materials and Methods 124

3.1. Reagents and antibodies .................................................................................... 124
3.2. Cell culture ........................................................................................................ 124
3.3. Fatty acid treatment ........................................................................................... 125
3.4. Sphingomyelinase assay .................................................................................... 125
3.5. Western blot analysis ......................................................................................... 125
3.6. Gene silencing of ASMase ................................................................................ 125
3.7. Isolation of caveolae/lipid rafts membrane microdomains ............................... 126
3.8. Isolation of plasma membrane domains ............................................................ 127
3.9. Lipid extraction ................................................................................................. 127
3.10. Mass spectrometry analysis ............................................................................. 128
3.11. Irnmunofluorescence assay .............................................................................. 129
3.12. Statistical analysis ........................................................................................... 130
4. Results 131
4.1. ASMase expression and activity in human retina] cells .................................... 131
4.2. ASMase cellular distribution in human retinal cells ......................................... 131
4.3. DHA effect on ASMase activity in human retinal cells .................................... 132
4.4. Downregulation of caveolar ASMase by DHA ................................................. 132

vii

4.5. Sphingomyelin enrichment of caveolae/lipid rafts by DHA ............................. 133
4.6. Gene silencing of ASMase prevents pro-inflammatory cytokine- induced HREC

 

activation .................................................................................................................. 134
5. Discussion 148
6. References 152

 

IV. DOCOSAHEXAENOIC ACID SUPPLEMENTATION PREVENTS
UPREGULATION OF PROIN F LAMMATORY/ ANGIOGENIC MOLECULES
AND VESSEL LOSS IN DIABETIC RETINA THROUGH DOWNREGULATION

 

 

 

 

 

 

OF ACID SPHINGOMYELINASE 158
1. Abstract 158
2. Introduction 160
3. Materials and Methods 163

3.1. Reagents and antibodies .................................................................................... 163
3.2. Rat model of type 1 diabetic retinopathy .......................................................... 163
3.3. The experimental diet ........................................................................................ 164
3.4. Isolation of retinal vasculature .......................................................................... 164
3.5. Histological assessement of retinal capillaries .................................................. 165
3.6. Simultanoeus extraction of proteins and RNA .................................................. 165
3.7. Quantitative real-time polymerase chain reaction ............................................. 166
3.8. Western blot ....................................................................................................... 167
3.9. Irnmunofluorescence assay ................................................................................ 167
3.10. Rat model of retinal ischemia-reperfusion ...................................................... 167
3.11. Slide preparation and quantification of ASMase staining of the blood vesselsl68
3.12. Statistical analysis ........................................................................................... 169
4. Results 170
4.1. DHA enriched diet prevents the upregulation of retinal ASMase in one month
diabetic rat model .................................................................................................. 170
4.2. DHA enriched diet prevents the upregulation of retinal
proinflammatory/angiogenic mediators in one month diabetic rat model ................ 171
4.3. DHA enriched diet prevents the upregulation of retinal ASMase in nine month
diabetic rat model ................................................................................................... 171
4.4. DHA enriched diet prevents the upregulation of retinal proinflammatory/
angiogenic mediators in nine month diabetic rat model ........................................... 172
4.5. DHA enriched diet prevents retinal vessel loss in nine month diabetic rat
model ........................................................................................................................ 172
4.6. Upregulation of ASMase and proinflammatory cytokines TNFa and IL-IB in
human diabetic retina ............................................................................................... 173
4.7. ASMase expression in the rat retinal vasculature .............................................. 173
4.8. DHA supplementation prevents the upregulation of vascular ASMase in the rat
retina injured by I/R .................................................................................................. 174
5. Discussion 191
6. References 195

 

V. ACID SPHINGOMYELINASE GENETIC DEFICIENCY PREVENTS
UPREGULATION OF INFLAMMATORY/ANGIOGENIC MEDIATORS,

viii

VESSEL LOSS AND PATHOLOGICAL NEOVASCULARIZATION IN THE

 

 

 

 

 

 

RETINA WITH MICROANGIOPATHY LESIONS 201
1. Abstract 201
2. Introduction 202
3. Materials and Methods 206

3.1 . Reagents ............................................................................................................ 206
3.2. Genotyping transgenic mice by PCR and Southemblot .................................... 206
3.3. Mice ................................................................................................................... 208
3.4. Isolation of retinal vasculature .......................................................................... 209
3.5. Histological assessement of retinal capillaries .................................................. 209
3.6. Simultanoeus extraction of proteins and RNA .................................................. 210
3.7. Quantitative real-time polymerase chain reaction ............................................. 210
3.8. Statistical analysis ............................................................................................. 210
4. Results 211
4.1. Retinal I/R induces upregulation of ASMase gene expression in the ASMase +/+
retina ......................................................................................................................... 211
4. 2. ASMase genetic deficiency prevents upregulation of cellular adhesion molcules
in the I/R injured retina ............................................................................................. 211
4. 3. ASMase genetic deficiency prevents upregulation of inflammatory/ angiogenic
mediators 1n the I/R injured retina ............................................................................ 212
4.4. ASMase genetic deficiency prevents the development of acellular capillaries in
the I/R injured retina ................................................................................................. 212
4.5. ASMase genetic deficiency prevents ocular neovascularization in OIR mouse
model ........................................................................................................................ 213
5. Discussion 221
6. References 226

 

ix

LIST OF TABLES

Table 4.1. Body weight gain and blood glucose of experimental animals with one of
month diabetes .......................................................................................................... 175

Table 4.2. Body weight at the end of the study, blood glucose and HbAl c of
experimenatal animals with nine months of diabetes ............................................... 175

Table 4.3. Fatty acid composition of the control diet and DHA (fish oil)- enriched

diet ............................................................................................................................ 176
Table 4.4. RT-PCR primers for rat ........................................................................... 177
Table 4.5. RT-PCR primers for human .................................................................... 177
Table 5.1 RT-PCR primers for mouse ...................................................................... 214

LIST OF FIGURES
Images in this thesis/dissertation are presented in color

Fig. 1.1. Fatty acids classification .............................................................................. 14
Fig. 1.2. Sphingolipids chemical structures ................................................................ 15
Fig. 1.3. Cerarnide generation by Sphingomyelinase-mediated sphingomyelin
hydrolysis ................................................................................................................... 35
Fig. 1.4. Receptor clustering and amplification of signal transduction by ceramide-
enriched macrodomains .............................................................................................. 36
Fig. 1.5. Diagram of working hypothesis ................................................................... 43
Fig. 2.1. Dose response of cytokine-induced CAMS expression in HREC ................ 85
Fig. 2.2. Cytokine induced ASMase and NSMase activation in HREC ..................... 86

Fig. 2.3. Inhibition of cytokine induced ASMase and NSMase activation, as well as
ASMase and NSMase expression by DHA in HREC ................................................ 88

Fig. 2.4. Inhibition of ASMase and NSMase cytokine-induced activation by DHA
and sphingomyelinases inhibitors ............................................................................... 90

Fig. 2.5. Decrease in TNF a induced CAMS expression by ASMase and NSMase

inhibition in HREC ................................................................................................. 91
Fig. 2.6. Inhibition in cytokine-induced CAMS expression by ASMase and NSMase

gene Silencing in HREC ............................................................................................. 93
Fig. 2.7. Inhibition of ASMase and NSMase activity by DHA .................................. 97

Fig. 2.8. No direct effect of DHA on sphingomyelinase (SMase) activity in a cell-
free system .................................................................................................................. 98

Fig. 2.9. Effect of cholesterol depletion on ASMase and NSMase activity in

HREC ......................................................................................................................... 99
Fig. 2.10. No effect of de novo ceramide production pathway inhibition on cytokine

induced CAMS expression in HREC ........................................................................ 102
Fig. 3.1. ASMase expression in human retinal cells ................................................ 135
Fig. 3.2. ASMase cellular distribution in human retinal cells .................................. 136

xi

Fig. 3.3. The effect of DHA on ASMase activity in HRPE and HMC .................... 139
Fig. 3.4. Downregulation of caveolar ASMase by DHA ..................................... 140

Fig. 3.5. The effect of DHA on sphingomyelin and ceramide content of caveolae
membrane microdomain in HREC ........................................................................... 141

Fig. 3.6. The effect of DHA and ASMase gene Silencing on IL-1 B-induced HREC
activation .................................................................................................................. 146

Fig. 4.1. The effect of DHA-enriched diet on ASMase retinal gene expression in rats
with one month of diabetes ....................................................................................... 178

Fig. 4.2. The effect of DHA-enriched diet on retinal inflammatory/angiogenic
molecules gene expression in rats with one month of diabetes ................................ 179

Fig. 4.3. The effect of DHA-enriched diet on retinal ASMase and VEGF protein
expression in rats with one month of diabetes .......................................................... 180

Fig. 4.4. The effect of DHA-enriched diet on retinal ASMase gene expression in rats
with nine months of diabetes .................................................................................... 181

Fig. 4.5. The effect of DHA-enriched diet on retinal ASMase protein level in rats
with nine months of diabetes .................................................................................... 182

Fig. 4.6. The effect of DHA-enriched diet on retinal inflammatory/angiogenic
molecules gene expression in rats with nine months of diabetes ............................. 183

Fig. 4.7. The effect of DHA-enriched diet on retinal vessel loss in rats with nine
months of diabetes ................................................................................................ 184

Fig. 4.8. Changes in retinal gene expression in control and diabetic human
retina ......................................................................................................................... 186

Fig. 4.9. Rat retinal ASMase expression ................................................................. 187
Fig. 4.10. Retinal vascular ASMase expression after retinal I/R injury in rats ....... 188
Fig. 5.1. Retinal gene expression of ASMase in the l/R retinopathy model ........... 215
Fig. 5.2. Retinal gene expression of NSMase in the I/R retinopathy model ........... 216

Fig. 5.3. Retinal gene expression of cellular adhesion molecules in the I/R
retinopathy model ..................................................................................................... 217

xii

Fig. 5.4. Retinal gene expression of inflammatory/angiogenic molecules in the I/R

retinopathy model .................................................................................................. 218
Fig. 5.5. Retinal capillary degeneration in retinal ischemia/reperfusion in I/R

retinopathy mouse model ......................................................................................... 219
Fig. 5.6. Retinal neovascularization in oxygen-induced retinopathy in mice .......... 222

xiii

AGE

ASMase

BSA

CAM

Cer

COX2

CRD

DD

DHA

DNA

DR

Elovl

FADD

FAN

HDL

HMC

HREC

HRPE

HSL

ICAM-l

IL

I/R

LIST OF ABBREVIATIONS
advanced glycation end product
acid sphingomyelinase
bovine serum albumin
cell adhesion molecule
ceramide
cyclooxigenase 2
caveolae-related domains
death domaain
docosahexaenoic acid (22:6n3)
deoxyribonucleic acid
diabetic retinopathy
elongase
FaS-associating protein with DD
factor associated with NSMase activation
high density lipoprotein
human Muller cells
human retinal endothelial cells
human retinal pigment epithelial cells
hormone—sensitive lipase
intercellular cell adhesion molecule -1
interleukin

ischemia/reperfusion

xiv

LAMP 1

LDL

LPL

MCD

MMP

MS

MUFA

NSMase

NF-KB

OIR

PKC

PCR

RNA

PUFA

ROS

SM

SMase

STZ

TRADD

VCAM-I

VEGF

VLDL

lysosomal associated protein 1
low density lipoprotein
lipoprotein lipase
methyl-B-cyclodextrin

matrix metalloprotease

mass Spectrometry
monounsaturated fatty acid
Neutral sphingomyelinase

nuclear-factor kappa B

oxygen-induced retinopathy
protein kinase C

polymerase chain reaction
ribonucleic acid

polyunsaturated fatty acid
reactive oxygen species
sphingomyelin

Sphingomyelinase

streptozotocin

tumor necrosis factor

TNF receptor associated DD
vascular cell adhesion molecule-1
vascular endothelial growth factor

very low density lipoprotein

XV

Chapter I
Literature Review

1. Diabetic Retinopathy

1.1. Overview

Current estimates suggest that approximately 135 million individuals across the
world have diabetes mellitus (1). In the US, over 23.6 million people are affected by
diabetes with Significant morbidity and mortality (2). Diabetic retinopathy (DR) stands as
the most common and disabling microvascular complication of diabetes and results in
blindness for over 12,000 people with diabetes each year (2). Moreover, DR is the
primary cause for blindness in the working age population worldwide (3) and in 2004,
vision loss from DR accounted for approximately $500 million in direct medical costs
among Americans age 40 or older (4). To place this in perspective, it is estimated that the
number of Americans 40 years or older with DR will triple during the period of time from

2005 to 2050, from 5.5 million in 2005 to 16.0 million in 2050 (5).

In conclusion, it is largely recognized that visual impairment among people with
I diabetic retinopathy is a major disability with a profound impact on patient independence
and can result in depression, reduced mobility and overall, poor quality of life (6).
Consequently, diabetic eye disease continues to represent an important socio-economic

burden on society and remains a challenge for the ophthalmology services provision.

1.2. Histopathological and Clinical Manifestations

Diabetic retinopathy represents the ocular expression of diabetic microvascular
complications. The natural history of the DR has been classified into an early, non-

proliferative stage (background retinopathy) and a late, proliferative stage.

The early, non-proliferative stage of DR has distinct vascular lesions that are
histologically characterized by the existence of saccular capillary microaneurysms, retinal
hemorrhage, venous dilation and beading, retinal lipid exudates, pericyte degeneration,
occlusion and degeneration of retinal capillaries (7-8). The mechanisms considered to be
involved in diabetic retinal vascular lesions include obstruction of the vascular lumen by
leukocytes or platelets; death of endothelial cells due to pathophysiological processes
within the vascular cells themselves; or endothelial cell death secondary to products
generated by other cells in the retina that are in close contact with these vascular cells

(such as neurons or glial cells) (7).

During the early phase of DR, clinical signs include retinal microaneurysms and
dot hemorrhages that are present in almost all patients with type 1 diabetes of a duration
20 years (9) and in approximate 80% of those affected by type 2 diabetes for the same
duration of time (10). As the disease progresses, the number and size of intraretinal
hemorrhages increases, and is accompanied by the appearance of cotton-wool spots on the
retinal background, indicative of ischemia due to local failure of the retinal microvascular
circulation. During the ophthalomoscopic investigation of the retina, retinal veins appear
dilated, tortuous and with irregular caliber, whereas arteries may appear white;

fluorescein angiography may reveal non-perfiised retinal arteries (11).

The late, proliferative stage of DR is histologically characterized by the presence
of aberrant new vessels, and fibrous tissue proliferation on the retinal surface that can
extend into the vitreous. The newly synthesized blood vessels are abnormal and fragile.
They tend to leak blood into the center of the eye inducing blurred vision and/or causing
blindness. As DR progresses, macular edema develops in the retina (due to the breakdown
of the blood-retinal barrier) resulting in leakage of plasma from the small retinal vessels

into the maculaand leading to severe loss of central vision (11).

The clinical signs of late stage DR are due to the proliferating retinal vessels along
with the glial tissue that anchor into the vitreous and lead to traction of the retina (3). In
the most severe cases, the fibrovascular tissue proliferation leads to total retinal
detachment and vitreous hemorrhage (12). Retinal neovascularization most commonly
takes place along the temporal vascular arcades and the optic nerve head. Retinal
neovascularization is an important contributor to the visual impairment and blindness in

diabetes.

Current approaches to prevention of DR involves general medical measures such
as control of blood glucose, blood pressure and serum lipids (11), whereas treatment of
DR involves specific therapeutic strategies that target retinal abnormalities encountered
during the course of the disease. The surgical options (laser panretinal photocoagulation
and vitreo-retinal surgical interventions) applied locally at the retinal level are highly

invasive, traumatic to the retinal tissue and do not prevent visual impairment.

Although the medical and surgical management of diabetes represent the mainstay

of DR management, they are not fiilly protective against visual impairment and patients

suffering blindness from diabetes are still arising, with approximately 8,000 new cases
reported annually. Therefore, new therapeutic approaches based on the etiology of early
diabetic microvascular lesions need to be developed and evaluated in clinical trials to

ultimately improve the outcome for patients with DR.

1.3. Epidemiology

Wisconsin Epidemiological Study of Diabetic Retinopathy (WESDR) was the
most comprehensive study of retinal neovascularization within a diabetic population,
where the prevalence and incidence of DR was investigated in more than 2000 patients
with diabetes (9-10, 13-14). This study showed that the prevalence of all retinopathy was
approximately 17% in patients with diabetes for less than 5 years, and 97.5% in patients
with diabetes for more than 15 years. The increase in prevalence of proliferative
retinopathy was strongly correlated with greater duration of diabetes and it varied from
1.2% in patients with diabetes mellitus for less than 10 years to more than 67% in patients
with diabetes for 35 years or more (9). This study on DR concluded that there is a strong
association between the frequency and severity of retinopathy and the duration of
diabetes. Moreover, the severity of DR was also related to a diagnosis at a younger age,
increased levels of glycosylated hemoglobin, the use of insulin, higher systolic blood
pressure, proteinuria and a small body mass (10).

New clinical data identified dyslipidemia as a critical factor in the development of
diabetic retinopathy. The Diabetes Control and Complication Trial

(DCCT)/Epidemiology of Diabetes Interventions and Complications (EDIC) identified a

strong association between the severity of DR in type 1 diabetes and lipid metabolism
(15). This study was conducted on patients with type 1diabetes and revealed a strong
association between severity of retinopathy in type 1 diabetes and the size of the particles
of three major classes of serum lipoproteins, very low density, low density and high
density lipoprotein (VLDL, LDL and HDL) as well as LDL concentration. Cross-
sectional studies Showed positive associations between the severity of retinopathy and
total- and LDL-cholesterol levels and the LDL-HDL cholesterol ratio (16). The Early
Treatment Diabetic Retinopathy Study has demonstrated that higher levels of serum lipids
are associated with an increased risk of development of hard exudates in the macula and
visual loss (17-18). Several studies Showed that lipid-lowering dietary (19) and drug (20)
therapy may lead to regression of retinal hard exudates, and that a diet high in

polyunsaturated fatty acids may protect against retinopathy (21).

2. Dyslipidemia and Diabetic Retinopathy

Dyslipidemia is not only a major metabolic disorder of diabetes mellitus, but also
a critical factor in the development of DR. Diabetic dyslipidemia with lipoprotein
abnormalities become manifested during the asymptomatic diabetic prodrome and

represent a high risk for vascular complications (22).

2.1. Insulin and Lipid Metabolism

Insulin, a polypeptide hormone secreted by B cells of the islets of Langerhans in
the pancreas, exerts an essential metabolic regulation not only on carbohydrate
metabolism, but also on lipid metabolism and protein synthesis. Insulin decreases
triacylglycerol degradation and hence the level of circulating fatty acids by inhibiting
hormone-sensitive lipase (HSL) activity in the adipose tissue (23). Moreover, insulin
increases glucose transport and metabolism in adipocytes and provides the substrate
glycerol 3-phosphate for triacylglycerol synthesis. In adipose tissue and muscle, insulin
also enhances the activity of lipoprotein lipase (LPL) and thus provides fatty acids for
esterification (24). LPL is then transported to the surface of endothelial cells where it
hydrolyzes the lipid chore of chylomicrons and low/very low density lipoprotein particles
and thus facilitate in the removal of triglyceride-rich lipoproteins from the blood stream.
Patients with either type 1 or type 2 diabetes have low adipose LPL activity, a higher rate
of HDL-metabolism and increase in plasma triglycerides level (25).

Fatty acid synthesis and remodeling is also under insulin control. Fatty acids are
classified according to their structure by the number of carbons, double bonds and

proximity of the first double bond to the methyl (-CH3) end of the fatty acyl chain.

According to the number of the double bonds, fatty acids are classified as saturated (no
double bonds), mono-unsaturated (MUFA, one double bond) and poly-unsaturated
(PUFA, multiple double bonds) and are presented in Figure 1.1. Saturated, mono-
unsaturated and poly-unsaturated fatty acids are synthesized from dietary precursors
(glucose, 16:0, 18:1n9, 18:2n6, 18:2n3, 18:2n6 and 20:5n3) through a series of elongation
(Elovl-2,—4,-5 and -6) and desaturation (AS—desaturase, A6-desaturase and A9-desaturase)
reactions. Insulin is the most potent activator of A5, A6 and A9 desaturase, where A5 and
A6 desaturases represent rate-limiting enzymes in fatty acid metabolism (25-30).
Therefore, low insulin availability or insulin resistance will result in impaired desaturase
expression, leading to substrate accumulation and product depletion. The major impact is
observed on long chain polyunsaturated a) 3 fatty acids that are no longer synthesized in
the desirable amounts. The lack of insulin in type 1 diabetes and insulin resistance in type
2 diabetes will promote a fatty acid profile with high saturation index and Short fatty acyl
chain and will lead to a Significant increase in HSL activity and decrease in PLP activity,

with a profound impact on plasma lipid levels, lipid profile and fatty acid composition.

2.2. Diabetic Dyslipidemia
2.2.1. Lipoproteins and Triglycerides

The natural course of development of type 2 diabetes (or insulin-resistant
diabetes) affects virtually all lipids and lipoproteins (31-32). The metabolic milieu is
characterized by accumulation of chylomicrons and VLDL remnants, triglyceride

enrichment of HDL and LDL particles, resulting in decrease level of HDL particles and

formation and accrual of small, dense LDL particles (33). Increased triglyceride
production along with decreased catabolism of triglyceride-enriched lipoproteins
heightens hypertriglyceridemia. Studies have shown that LDL particles become small and
dense in approximatively 90% of the patients whom triglycerides levels are > 200 mg/dL;
when triglyceride levels are shifted 90 mg/dL, almost all LDL lipoprotein particles are of
the larger, buoyant range (34).

The epidemiological data on the prevalence of dyslipidemia and phenotype
distribution showed that, regardless of a poor glycemic control, total triglyceride and
cholesterol levels are the same when compared patients with insulin-dependent type 1
diabetes and nondiabetic, control, healthy subjects age matched (35). However, increased
LDL-cholesterol levels and decrease in HDL-cholesterol have been found in diabetic
group compared to nondiabetic group (3 5). More importantly, the lipid abnormalities may
improve with a good glycemic control, except for HDL-cholesterol levels (35).
Kordonouri et al. also Showed that HDL-cholesterol is the most important variable,
tightly correlated to the development of retinal lesions in children with type 1 diabetes

(36).

2.2.2. Long chain PUFAS

Over the past few years, there has been an increasing interest in exploring the role
of long chain PUFAS in pathophysiology of diabetic metabolic complications. These
PUFAS contains more than one cis double bond and are classified as (06 or (133-PUFAS
according to the location of the first double bond compared to the fatty acid methyl end as

shown in Figure 1.1. (03- PUFAS are regarded as important anti-inflammatory mediators

(37-39), whereas (06- PUFAS are considered to have pro-inflammatory properties (40).
More importantly, diabetic dyslipidemia is characterized by an elevated (06 to w3-PUFA
acid ratio (41-42) that can "incline the body physiologic equilibrium" toward an
inflammatory state. Ferruci et al. studied the relationship between plasma PUFAS and
circulating inflammatory markers in a community based-sample and found that increased
total (0-3 PUFAS levels were associated with lower levels of inflammatory markers (IL-6,
IL-lra, TNFa, C reactive protein) and higher levels of anti-inflammatory markers (soluble
IL-6r, IL-lO) (43-44). The authors concluded that (D3-PUFAS are beneficial in patients

affected by diseases with an active inflammatory component.

2.2.3. Sphingolipids

New research revealed the importance of another class of lipids that are
dysregulated in diabetes, the sphingolipids. Sphingolipids represent a complex class of
lipids and the prefix “sphingo-” that appears in the nomenclature of this lipid class was
first attributed by J.L.W. Thudichum in 1884 because the enigmatic nature of these
molecules reminded him of the riddle of the Sphinx. In 1974, Herbert Carter and
colleagues introduced the “sphingolipide” name. Sphingolipids are extremely versatile
molecules. Relative to phospholipids, sphingolipids are enriched in saturated long fatty
acyl chains, are more hydrophobic, have a higher melting temperature and can undergo a
tighter packing in the cellular membrane due to the lack of unsaturated acyl chains with
kinked structures (Figure 1.2) (45-46). Sphingolipids have a chemical structure composed
of a 1,3-dihydroxy—2-aminoalkane backbone, also known as the sphingoid base.

Sphingosine, the most prevalent backbone of mammalian sphingolipids is characterized

by a (28,3R,4E)-2-amino-4-octadecene-1,3-diol chemical structure. Saturation of the
sphingosine backbone results in sphinganine, which represents another sphingoid base
frequently used to form mammalian sphingolipids. Attachment of a fatty acid to the
sphingosine backbone via an amide bond results in formation of ceramides; attachment of
a variety of polar headgroups to the ceramide structure results in formation of complex
sphingolipids, such as Sphingomyelin, gangliosides, sulfatides, globosides or
cerebrosides.

Sphingolipids such as ceramide and glucosylcerarnide, while representing a
relatively minor component of the lipid milieu in most tissues, may be among the most
pathogenic lipids in the onset of diabetic metabolic complications, including insulin
resistance, pancreatic B-cell failure and vascular dysfunction (47-48). Circulating factors
associated with diabetes (e. g. saturated free fatty acids, inflammatory cytokines)
selectively induce enzymes that promote ceramide synthesis and stimulate the
accumulation of ceramide and various ceramide metabolites (49-52); these sphingolipids
have been Shown to amass in tissues from rodents (53-54) and humans (55-56) with type
2 diabetes. Extensive lipidomic profiling revealed ceramide accumulation in the muscle
tissue from rats on a high fat diet for three weeks or from streptozotocin diabetic rats, one
month after diabetes induction (54). Related to these findings, Straczkowski et al. have
shown elevated ceramide levels in the skeletal muscle of obese men at risk of developing
type 2 diabetes (57). Interestingly, a fish oil (DHA) enriched diet significantly reduced
ceramide content in the mouse muscle tissue (58).

Dysregulation of sphingolipid metabolism in diabetes emerges as an important

pathway for development of the deleterious clinical manifestations associated with this

10

disease. Importantly, sphingolipid metabolism was found to be Significantly altered in
diabetic retina (59). Therefore, pharmacologic inhibition or genetic ablation of key
regulatory enzymes of sphingolipid metabolism may identify new therapeutic targets for
combating insulin resistance, B-cell failure and vascular dysfunction of diabetic metabolic

disease.

2.3. Retinal Fatty Acid Profile and Diabetic Retinopathy

The normal retina has a very specific fatty acid profile, with the highest level of
long-chain PUFAS in the body. Various studies have identified DHA, an (0-3 PUFA as
being enriched in the retinal tissue (60-61). Work from our group demonstrated that DHA
represents 45.37 d: 1.32 mole % of total fatty acids identified in the retina (61). Omega 3-
PUFAS and particularly DHA have the ability to modulate various biological processes
involved in retinal vascular inflammation, retinal capillary structure and integrity
alterations, and retinal neovascularization (62). Therefore, (03 long-chain PUFAS and
mainly DHA status in the retina is closely related to normal retinal structure and function

(62).

Long chain PUF A synthesis and remodeling involves a series of desaturation (A5-
desaturase, A6-desaturase, A9-desaturase) and elongation (Elov11-7) reactions. The
elongases Elov12 and Elov16 are ubiquitously present in most tissues; yet, the retinal
tissue express very high levels of Elov12 and this elongase is involved in several steps of
DHA synthesis (61, 63). Elovl4 has a restricted tissue specificity, being highly expressed
in the retina (64-66), thymus and skin (65), and to a lesser extent in the brain (65-66) and

testis (66). Insulin deficiency or complete absence of insulin downregulates Elov12 and

11

Elovl4 elongases Specifically in the retina, and thus precludes the synthesis of various
retinal PUFAS, and particularly (D3-PUFAS and DHA (61). We have found that retinas
isolated from type 1 diabetic animals had a 28% reduction in DHA levels when compared
to control (61). Previous studies have also demonstrated decreased levels of oo3-PUFAS,
and mainly DHA in plasma of diabetic children (67), as well as in human retina of
diabetic eyes (68). More importantly, decreased retinal DHA levels, and an increase in
the w6-to- (o3 PUFA ratio in diabetic retina was associated with significant upregulation
of [CAM-l and inflammatory cytokine (VEGF, TNFa and IL-6) gene expression,
suggesting that DHA tissue insufficiency may create pro-inflammatory conditions in the
retina, potentially contributing to the development of DR (61). Our study also
demonstrates that retinal fatty acid metabolism is distinct from the metabolic changes that

occur in diabetic plasma or diabetic liver (61 ).

Several studies have identified DHA retinal insufficiency to be associated with
alterations in retinal function (reviewed in (62)) and DHA deficiency has also been
documented to be associated with certain retinal degenerative diseases, such as retinitis
pigrnentosa, retinopathy of prematurity and age-related macular degeneration (69-73).
Taken together, all these demonstrate that DHA has a critical role in maintaining the

normal structure and function of the retina.

A study by Connor et al. showed a beneficial effect of dietary DHA in reducing
pathological retinal angiogenesis and thus preventing the development of retinopathy (74)
in an oxygen-induced retinopathy (retinopathy of prematurity) animal model.
Furthermore, it has been Shown that DHA supplementation could ameliorate visual

processing deficits (62). In vitro studies from our group demonstrated a significant

12

protective effect of DHA against cytokine induced inflammation in human retinal
endothelial cells (39-40). However, the protective effect of DHA against diabetic retinal
inflammation and microangiopathic lesions has not been addressed and represents an

objective of my study.

13

Saturated Fatty Acids 0

Palmitic Acid (16:0) HOW/WC”:

Mono Unsaturated Fatty Acids

HO _
WWW.

Oleic Acid (18:1,n9) O

Poly Unsaturated Fatty Acids HO N6

Linoleic Acid (18:2,n6) (n6-PUFA) OWMH3

Poly Unsaturated Fatty Acids
COOH

Docosahexaenoic Acid
CH
(DHA, 22:6,n3) (n3-PUFA) N3 3

Figure 1.1. Fatty acids classification. Fatty acids classification is based on the number
of carbons in the chain length, the number of double bonds and the localization of the first
double bond related to the methyl end. Based on these considerations, fatty acids can be
classified as saturated, monounsaturated and polyunsaturated.

14

Sphingoid bases

Sphingosine OH

/_/I\/CH20H
WM _ 5

NHz

Dihydrosphingosine OH
WNk/CHZOH

NH2

Phytosphin osine
9 OH

xc\\,#“\c/”\c//‘xa/“\sx”\cArbxg/Lxc/CHon

OH NHg

OH

Dehydrophytosphingosine
/\/\/\/\/—/\/\/i\/ CHZOH

OH NHg

Figure 1.2. Sphingolipids chemical structures. Sphingoid bases are long-chain aliphatic
amines, with two or three hydroxyl groups, and ofien a characteristic trans-double bond
in position 4; they are 2-amino-1,3-dihydroxy-alkanes or alkenes with (2S,3R)-erythro
stereochemistry, with various structural modifications.

15

Ceramide

Sphingomyelin

Figure 1.2 (continued). Sphingolipids chemical structures. Ceramide consists of a
sphingoid base linked to a fatty acid via an amide bond. Sphingomyelin consists of a
ceramide molecule with a phosphorylcholine moiety attached to position 1.

16

3. Inflammatory Mechanisms in Diabetic Retinopathy

An extensive body of evidence, including both preclinical and clinical findings,
supports the concept that very early stage diabetic retinopathy represents a persistent low-
grade chronic inflammatory disease that involves increased expression of adhesion
molecules on endothelial cell surfaces, eventuating in leukocyte recruitment, retinal
cytokine production and activation of transcription factors that result in up-regulation of
retinal inflammatory genes. These pathophysiological changes will be separately

discussed and evidence to support their role in DR will be presented.

3.1. Adhesion Molecules in Diabetic Retinopathy

The first line of evidence supporting a connection between inflammation and
diabetic retinopathy came from an epidemiological data in 1964 Showing a decrease in
severity of DR in diabetic patients treated with high doses of aspirin for rheumatoid
arthritis (75). In a rat model of diabetic retinopathy, aspirin and meloxicam (nonsteroidal
anti-inflammatory agents) prevented the development of early diabetic retinopathy, a
result that correlated with suppress of intercellular adhesion molecule-1 (ICAM-1)
expression and leukocyte adhesion (76). Adhesion molecules, especially ICAM-1 and
vascular cell adhesion molecule-1 (VCAM-l), are involved in leukocyte attachment and
transmigration into the vascular intima (77-78) that could result in early blood retinal
barrier breakdown, capillary non-perfusion and endothelial cell injury and death.
Evidence supporting this view are provided by findings of Significant increase in
leukocyte density and retinal vascular ICAM-l expression in human eyes with DR (78).

Moreover, a role of CD18/ICAM-l-mediated leukocyte adhesion in the pathogenesis of

17

early diabetes-induced leukostasis and blood-retinal barrier breakdown has been
established. Reduced expression of ICAM-1 on endothelial cells or its leukocyte counter-
receptor CD18 Significantly inhibits leukocyte adhesion, pericyte and endothelial cell loss
and consequently the formation of acellular, degenerated capillaries in experimental
models of diabetic retinopathy (79). Theoretically, an intervention at this level might

carry an important role in designing future treatment strategies in DR.

3.2. Proinflammatory Cytokines in Diabetic Retinopathy

Proinflammatory cytokines (tumor necrosis factor alpha (TNFa), interleukin 1
beta (IL-10), and vascular endothelial growth factor (VEGF) are increased in diabetic
eyes and several inflammatory pathways are activated in the very early stages of DR (79-
82).

TNFa is a proinflammatory cytokine implicated in a number of inflammatory
diseases, including rheumatoid arthritis, Chron's disease, ankylosing spondylitis and
psoriasis (83). In diabetes, TNFa has been involved in pancreatic B cell apoptosis (84) and
insulin resistance (85), as well as in diabetic microvascular complications, including
nephropathy (86) and diabetic retinopathy (76). TNFa acts through ligand-receptor
interactions and initiate a number of inflammatory processes, such as upregulation of
adhesion molecules expression, leukocyte recruitment and leukostasis, cellular apoptosis,
chemoattraction of monocytes and plays a critical role in amplification of the immune
reSponse through upregulation of the expression of numerous transcription factors, growth

factors and other inflammatory mediators (86). Clinical and preclinical studies indicate an

important role of TNFa in the pathogenesis of DR. Correlative clinical studies with

18

genetic data have implicated a TNFa polymorphism in confen'ing susceptibility to the
disease (87). Moreover, retinal levels of TNFa are significantly increased in diabetic rats
compared to control (76) and increased amounts of TNFa can be found in the
extracellular matrix, vessel walls, endothelium and vitreous of diabetic eyes affected by
proliferative retinopathy (80, 88-89). Playing a central role in the pathogenesis of DR,
TNFa has been targeted with specific medication aimed to counteract its inflammatory
effects.

Systemic administration of etemacept, a soluble TNFa receptor/Fe fusion protein
that acts as a competitive inhibitor to block the effects of TNFa on cells, resulted in
reduction of the inflammatory process in DR, including leukocytes adhesion to the retinal
vasculature, upregulation of ICAM-1, NF-KB activation and breakdown of the blood
retinal barrier (76). The effects of etemacept were not evaluated on advanced histologic
lesions in DR, but it was reported that mice genetically deficient in TNFa were protected
against galactose-induced retinopathy (7). Administration of lnfliximab, a monoclonal
antibody directed against TNFa, for two months in patients with advanced DR Showed
positive responses, with improvement of visual acuity and reduction in macular edema
and further improvements were seen after repeated infusions (90).

Increased intraocular VEGF expression, a growth factor recently recognized to
have cytokine properties, has been identified in rat and human diabetic retina (82, 91-94)
and has become a focal point of current research on the pathogenesis of DR. The VEGFS

are a family of peptides produced by alternative splicing fiom a single gene; VEGF
1'80me3 are particularly mitogenic for vascular endothelium and also amplify

permeability at blood-tissue barriers (1 1). VEGF is synthesized by multiple cell types in

19

the retina, such as ganglion cells, Muller cells and pericytes (95-97). VEGF induces
upregulation of lCAM-l expression in endothelial cells (98), resulting in chronic retinal
leukostasis that plays a key role in endothelial cell injury and death and blood-retinal
barrier breakdown (94). VEGF is a key molecule that plays a well-recognized role in
promoting vascular permeability (99) and pathological ocular neovascularization (100).
VEGF expression is mainly regulated by hypoxic stimulus; yet there is evidences that it
also accumulates in the retina early in diabetes, before any retinal hypoxia is present (101 -
102). Therefore, VEGF upregulation is strongly involved in the pathogenesis of both non-
proliferative (background) and proliferative DR (81-82, 94, 103). Specific inhibition of
VEGF hinders ICAM-l upregulation, leukostasis, blood-retinal barrier breakdown and
pathological ocular neovascularization in diabetic rats (98). Clinical trials using anti-
VEGF therapeutic strategies intravitreously administered (pegaptanib, a VEGF inhibitor;
ranibizumab and bevacizumab, monoclonal antibodies that bind all VEGF isoforms)
showed promising results against advanced stages of DR (104-106). Among these,
ranibizumab is the only anti-VEGF therapy that was approved by Food and Drug
Administration for ophthalmologic use in neovascular (wet) age-related macular
degeneration; yet, it's efficacy in DR remains to be investigated.

Pro-inflammatory cytokine IL-lB levels are also increased in retinas from diabetic
rats (107-108). In vivo studies with intravitreal injection of IL-IB or exposure of retinal
endothelial cells to IL-lB in vitro was able to induce endothelial cell inflammation and
death, resulting in degeneration of retinal vascular endothelium and formation of acellular

capillaries (109). Caspase-l, the enzyme involved in production of proinflammatory

cytokine IL-lB from pro-IL-l B, is activated in the retina of diabetic patients and diabetic

20

animal models (110). The involvement of IL-lB in pathogenesis of DR has been recently
defined in diabetic mice in which caspase-1 was inhibited or the IL-IB receptor was
deleted (108). Inhibition of caspase-1 using minocycline or inhibition of IL-lB signaling
using IL-IB receptor genetic deficient mice protected the mice from development of
retinal pathology (108). These data suggests a major role for IL-IB and its receptor in
mediating signaling pathways responsible for endothelial cell activation and death, and
consequent retinal vessel loss.

Proinflammatory cytokines that are upregulated in diabetic retina (i.e. TNFa, IL-
10 and VEGF) are likely to activate nuclear factor-kappa B (NF-KB) pathway (40). NF-
KB proteins consist of a family of structurally related transcription factors widely
expressed in mammalian tissues, which control a large number of cellular processes, such
as immune and inflammatory responses, developmental processes, cellular growth, and
apoptosis (7). NF-IcB transcription factors represent a prerequisite for various gene
expressions, including adhesion molecules, inflammatory cytokines, cyclooxygenase 2
(COX2) and matrix metalloproteases (MMPs) (111-112). These transcription factors are
persistently active in various disease states, including chronic inflammation, arthritis,
asthma, cancer, and heart disease (113). DR is hypothesized to be a chronic inflammatory
disease and activation of NF-KB is well documented in the retinal vasculature of diabetic

patients and animals (114-115).

TNFa and IL-lB have been shown to activate cellular sphingomyelinases in
various cell types (116-119). Stimulation of the sphingomyelin pathway by TNFa,
andSphingomyelinases-induced membrane ceramide formation leads to activation of NF-

KB and marked increase in nuclear NF -I<B localization in human leukemia (HL-60) cells

21

(118). In the same way, activation of Sphingomyelinases is shown to be an important
signaling system for induction of IL-10 in the murine T helper cells EL-4 (1 1, 7); IL-lfl
action in these cells is mediated through NF-KB activation (120). Sphingomyelinases may
play an important function in mediating the inflammatory process in DR; their role in

vascular retinal inflammation has not been addressed and represents the objective of my

study.

22

4. Sphingomyelinases and Inflammation

Sphingomyelinases (sphingomyelin phosphodiesterase) are key regulatory
enzymes in sphingolipid metabolism. They induce sphingomyelin hydrolysis to generate
the pro-inflammatory and pro-apoptotic second messenger ceramide. Several isofonns of
sphingomyelinases have been identified and further distinguished by their catalytic pH
optimum, cellular localization, primary structure and co-factor dependence. Alkaline
sphingomyelinase activity is confined to the intestinal mucosa, bile and liver and does not
participate in signal transduction (121-123). Neutral (NSMase) and acid (ASMase)
sphingomyelinases, however, are crucially involved in pathophysiology of metabolic

disorders (47) and play an active role in cellular signaling (124).

4.1. Neutral Sphingomyelinase

Neutral sphingomyelinase (N SMase) activity was described for the first time by
Schneider and Kennedy in 1967 when they found that spleens of patients with Niemann-
Pick disease (deficiency of ASMase) contained a magnesium-dependent enzyme capable
of promoting sphingomyelin hydrolysis to ceramide at an optimum pH of 7.4 (125).
NSMase shows an absolute dependence on magnesium or manganese for its activation,
with a best concentration ranging between 2 and 10 mM for magnesium and up to 5 mM
for manganese (126-127). In vitro studies showed that NSMase requires neutral

detergents for presentation of its substrate in mixed micelles (127).
NSMase has an estimated molecular mass of 92 kDa, although smaller
isoenzymes have been described, which could result from either the proteolytic process or

from an alternative processing of common transcripts (128). Subsequent research showed

23

NSMase activation can occur in response to a variety of stimuli, such as cytokines,
cellular stress (UV light, chemotherapeutic drugs), amyloid B and lipopolysaccharide
(124, 129). Activation of NSMase by such a wide array of stimuli further highlights the
significance of this enzyme as an important regulator of ceramide production and
ceramide-dependant Si gnaling (130).

NSMase appears to be ubiquitously expressed in mammalian tissues, with a
particular high activity in brain (128) and to a lesser extent, Spleen and liver (130).
NSMase has been reported to be localized to the plasma membrane in mammalian cells
(128), has two putative transmembrane domains at the NH; terminus and palmitoylated on
five cysteine residues via thioester bonds and it likely has affinity for caveolae/lipid rafts
microdomains (124, 131-132). Moreover, Veldman et al. described a NSMase in caveolae
isolated from human skin fibroblasts where the enzyme appeared to interact with the
scaffolding domain of caveolin protein (133). Also, Czamy et al. have reported a caveolar
form of NSMase in the caveolin-enriched membrane fractions isolated from bovine lung
microvascular endothelial cells (134).

Programmed cell death is an essential and strongly regulated process for normal
tissue development and homeostasis. The role of ceramide as a major mediator of
apoptosis is well-established. Data in the literature indicate sphingomyelinases-induced
ceramide production as a major pathway in this process, a theory supported by the
activation of NSMase in response to various apoptotic stimuli, including cytokines, serum

starvation and heat stress (135-136). Several studies have described a possible role for
NSMase-mediated ceramide production in regulating cell cycle and growth arrest (137-

138)- There exist strong evidence for sphingolipids as mediators of inflammatory

24

responses such as prostaglandin production, upregulation of adhesion molecules and
leukocyte recruitment (139) and the possibility has been raised that NSMase activation
could be involved in the cytokine-induced inflammatory process (137).

TNFa and IL-lB activate cellular sphingomyelinases in different cell types (116-
119). The mechanisms by which IL-IB activates NSMase remains to be elucidated.
Majority of studies on Sphingomyelinases activation have used a model system utilizing
the p55 TNFa receptor. Kronke et al. have defined a region of p55 receptor that is
adjacent to the death domain and is required for NSMase activation (50). This motif was
named the NSMase activation domain. The same authors have also described a novel
protein (factor associated with NSMase activation, FAN) that binds the NSMase
activation domain on the TNFa receptor and is required for NSMase activation (140).
Overexpression of full-length FAN increased NSMase activity in TNFa treated cells,
whereas a truncated mutant of FAN exerted dominant negative effects. Similarly, a
dominant negative form of FAN significantly decreased cell death induced by CD40 and
TNFa in transformed human fibroblasts (141-142). Although all these studies begin to
shed some light on NSMase mechanism of activation, the mechanisms coupling FAN to

NSMase have not been determined.

4.2. Acid Sphingomyelinase
Acid sphingomyelinase (ASMase) is a soluble glycoprotein and the first described
Sphingomyelinase. The complete absence of this enzyme is responsible for Niemann-Pick
Syndrome, a major neurological disorder (128). The gene encoding ASMase is designated

as Sphingomyelin phosphodiesterase 1 (SMPDI), is located within an "imprinted"region

25

of the human genome (chromosomal region 1 lp15.4) and is preferentially expressed from
the maternal chromosome. This form of genetic regulation is classical. for genes that play
a critical role in development ( 143-144).

The biosynthesis of ASMase begins with a 75 kDa pro-precursor that is
proteolitically cleaved to a 72 kDa precursor in the endoplasmic reticulum and Golgi. The
72 kDa form is then processed in the endosome-lysosome compartment to a 70 kDa
mature and firlly active lysosomal ASMase. The mature forms of ASMase, as well as its
precursors, possess six potential N-glycosylation sites and out of this, at least five are
critical for proper folding, protection against proteolysis and enzymatic activity (145-

146).

4.2.1. Tissue Distribution and Cellular Localization

ASMase is a Sphingomyelin phosphodiesterase enzyme ubiquitously express in
various mammalian tissues (128) and is involved in various metabolic processes and
Signal transduction pathways. However, vascular endothelium is well recognized to
express 20 times as much ASMase as any other cell type in the body (147) and it has
been shown that this large surplus of ASMase in endothelium may be involved in tissue
remodeling and wound repair (148), processes that normally require waves of endothelial

cell apoptosis and tissue neovascularization (149).

Although ASMase was initially described as a strictly lysosomal enzyme because
01‘ its optimum activity at pH 4.5-5.0, an ASMase isoforrn was recently shown to be

localized into secretory vesicles in close proximity to the plasma membrane and to be

26

secreted extracellularly upon cell stimulation (150); moreover, ASMase activity has also

been observed in caveolae/lipid rafts membrane microdomains.

The secreted form of ASMase is stimulated by zinc, whereas the intracellular,
lysosomal isoforrn is already tight bound to this cation; therefore, both ASMase isoforms
require zinc for their activity (151). The secretory and lysosomal forms of ASMase are
encoded by the same gene and display differences in the oligosaccharide structure as well
as in the N-terrninal proteolytic processing (151-152). The lysosomal form is rich in
mannose-type oligosaccharides which are phosphorylated for lysosomal targeting; the
secretory form of ASMase has complex-type, N-linked oligosaccharides (15]).

Liu and Anderson have described the plasma membrane form of ASMase present
in caveolae and they have shown that lL-IB stimulates its activity in this compartment
(116). The same authors were first to report that ASMase localize to the caveolae
microdomains in human fibroblasts and its activity was zinc-independent, suggesting that
this ASMase was not the zinc-dependent secretory form of ASMase. Another study
brought out evidence of the existence of a small pool of zinc-independent ASMase in the
plasma membrane microdomains purified from NIH-3T3 L1 fibroblasts and PC12 cells,
and more importantly, this pool of ASMase could undergo activation in response to
neurotrophins, whereas the lysosomal ASMase did not respond to activation by this
ligand (153).

Various immunohistochemistry studies have Shown that cells exposed to different
stress stimuli exhibit a change in ASMase location from intracellular compartment to the
cell surface (154). For instance, ASMase translocates onto the extracellular leaflet of

plasma membrane upon activation, a process that is assumed to be mediated by the fusion

27

of the vesicles containing ASMase with the plasma membrane (155). It was shown that
this process occurs within seconds after cellular stimulation, for example after CD95
ligation (150, 156). A recent study described ASMase activation and translocation to the
outer leaflet of the plasma membrane by a mechanism that involves ASMase
phosphorylation at serine 508 by protein kinase C5 (PKCS) upon cellular stimulation with
phorbol ester (157). Zeidan et al. demonstrated that activated PKCB is translocated to
vesicles containing ASMase, PKC6 then phosphorylates ASMase; the phosphorylated
enzyme then trafficks to the extracellular leaflet of the cell membrane where it generates

ceramide and promotes the formation of ceramide-rich membrane platforms.

4.2.2. Mechanisms of Regulation

The mechanisms that control ASMase activation are not yet completely
elucidated; however, several factors have been described to regulate ASMase activity in
vitro and in vivo settings.

pH and ions. AS previously mentioned, the optimum pH for ASMase activity
ranges between 4.5 and 5.5; however, a pH increase appears to affect only substrate
affinity and not ASMase activity (158). Although ASMase present at the cell surface
functions best at acid pH, it is also capable of hydrolyzing certain physiologic substrates,
including caveolae/lipid raft associated sphingomyelin or atherogenic proteins, at neutral
or slightly acidic pH (159-160). Divalent cations (cobalt, calcium, magnesium and
manganese) up to 250 mM and EDTA and EGTA have no effect on ASMase activity;

however, the enzyme requires tightly bound zinc for its activity (124).

28

Lipids. It has been shown that various lysosomally occurring lipids, including his
(monoacylglycero) phosphate and phosphatidylinositol were highly effective in ASMase
activation (124). Several reports postulated 1,2 diacylglycerol (DAG)-mediated ASMase
activation upon cell stimulation with pro-inflammatory cytokines (TNFa, IL-IB) (116,
161-162) or Fas receptors (163). A recent study showed ASMase inhibition by
sphingosine-l-phosphate in macrophages isolated from apoptotic bone marrow (164).

However, the mechanisms of ASMase regulation are not completely understood.

Proinflammatory cytokines. Pro-inflammatory cytokines that are increased in
diabetic eyes, TNFa and IL-IB have been demonstrated to activate cellular
sphingomyelinases in various cell types (116-119). The mechanisms by which IL-lB
stimulates ASMase activity have not yet been elucidated. p55 TNFa receptor was used as
model system to study sphingomyelinases activation, showing that this receptor has
particular domains for different sphingomyelinases. The intracellular C-tenninal part of
the p55 receptor, which contains a 75-amino acid motif, termed the death domain (DD),
appears to be necessary for ASMase activation; the exact molecular mechanism is not
known, but Schwandner et al. have reported ASMase activation (seconds to minutes) in
response to TNFa through overexpression of TRADD (TNF receptor associated DD)

and/or FADD (Pas-associating protein with DD) (165).

Other regulatory factors. Several studies have described a role for PI-kinase
pathway (166), tumor suppressor p53 (167) and oxidants (168-169) in regulation of
ASMase activity, but further studies are needed to elucidate the relationship between

these pathways and ASMase activation.

29

4.2.3. Roles of ASMase

ASMase is considered to be an important mediator of inflammation and apoptotic
effects due to various stimuli, including radiation (170-171), chemotherapy (172),
ischemia (173), TNFa (174), Fas/CD95 (175). Lysosomal ASMase has a classical,
important role is Sphingomyelin storage and metabolism, as shown by Sphingomyelin
accumulation within the lysosomal compartment in Niemann-Pick disease (ASMase
deficiency) (176). However, it is now believed that ASMase plays a key role in initiating
cell signal transduction at the cell membrane level (and not in the intracellular
compartment) (177) where it participates in formation of ceramide-enriched membrane
microdomains that coalesce to form ceramide-enriched platforms, receptor clustering and

amplification of Signal transduction via a specific receptor (124, 178-180).

4.3. Sphingomyelinases, Plasma Membrane Microdomains and Inflammatory Signal
Transduction
4.3.1. Plasma Membrane Microdomains

4.3.1.1. Classification and functions

Over the past decade, a number of membrane fractions could be isolated by their
insolubility in the non-ionic detergent Triton X-100 at 4°C, conditions under which most
cell membranes are disrupted. Due to their high lipid content, these detergent-resistant
membrane fractions can be further isolated by flotation in sucrose gradients and density
gradient centrifugation, due to their relatively low density (181). The detergent-resistant

fractions enclose at least two kinds of structures: a) caveolae, bottle-shaped invaginations

30

of the plasma membrane that contain a cholesterol-binding protein named caveolin-land
b) non-caveolin containing membranous structures, also known as "rafts" or caveolae-
related domains (CRD). Retinal endothelial cells contain both lipid rafts and caveolae. In
spite of initial doubts, strong evidence supports the hypothesis that detergent-resistant,
lipid-enriched domains do exist in cell plasma membrane in vivo and are not an artifact of
detergent extraction (181-183). In addition, caveolae structures have been visualized in
vascular endothelial cells since 1953 as 50500 nm cell surface invaginations (184) and
caveolae fiactions have been isolated and purified in the absence of detergents (185).

Caveolae were initially believed to play an important role in regulation of
transcytosis in endothelial and epithelial cells (186). However, at present, a more complex
role for caveolae has been revealed, including regulation of vascular permeability (187),
lipid trafficking, cholesterol homeostasis (188-189) and, in particular, signal transduction
(190-191).

Caveolae are stabilized by particular membrane spanning proteins, the caveolins, a
family of palmitoylated hairpin-like membrane proteins, which organize Q-shaped cell-
surface invaginations (192-193). Caveolin-l, the primary structural protein constituent of
caveolae, is a biochemical marker for establishing the putative existence of caveolae
(193).

Both types of detergent-resistant membrane fractions are enriched in cholesterol,
sphingolipids and glycerophospholipids (194). The lipid core seems to play an important
role in attracting lipid-modified membrane proteins to caveolae; many peripheral and
integral molecules associated with caveolae/lipid rafts are dually lipidated with saturated

acyl chains that readily group into this ordered lipid enviromnent (195-197). The acyl

31

chains of these proteins intercalate in the lipid bilayer, collect in caveolae due to a slow
lateral mobility (upon encountering the liquid-ordered phase that characterize the
caveolae) and determine protein-protein and protein-lipid interaction at this Site (198-
200). The membrane surrounding the caveolae/lipid rafts is more fluid since is mainly

composed of unsaturated phospholipids.

4.3.1.2. Sphingolipids and Plasma Membrane Microdomains

Sphingolipids represent a major component of the plasma membrane
microdomains (122). Compared to phospholipids, sphingolipids have saturated long fatty
acyl chains in their structure and can undergo a tighter packing in the cellular membrane
due to the lack of unsaturated acyl chains with kinked structures (Figure 1.2) (45-46). In
these membrane microdomain fractions, sphingolipids interact with each other via
hydrophilic interactions between the sphingolipid headgroups (201-202). Moreover,
Sphingomyelin, the predominant sphingolipid in these membrane microdomains, interacts
with cholesterol via hydrogen bonds with the hydroxyl group in the cholesterol structure
and via hydrophobic van der Waal interactions between the ceramide moiety and sterol
ring structure (155). This provides an extra element of stability of these membrane
structures, which are said to exist in the "liquid-ordered state".

Sphingolipids may have vital roles in caveolae/lipid rafts structure and are now
known to act as messengers in signaling pathways mediating inflammation, apoptosis,
cell differentiation and proliferation (203). Lipid analysis showed that approximately 70%
of total cellular sphingomyelin is found in caveolae/lipid rafts domains (122, 202). New

evidence suggests that enrichment in sphingomyelin content in caveolae/lipid rafts make

32

them potential substrate pools for cellular sphingomyelinases to produce a high local

concentration of ceramide.

4.3.2. Sphingomyelinases and Plasma Membrane Microdomains

Caveolae/lipid rafts appear to be definite sites for ceramide production in response
to different agonists and stress signals. Activation of various forms of sphingomyelinases
can hydrolyze the phosphodiester bond of sphingomyelin to ceramide and
phosporylcholine (Figure 1.3) (122). Both NSMase and ASMase are rapidly and
transiently activated by various exogenous stimuli, leading to a significant increase in
ceramide levels in a time frame of seconds to minutes (204-206). Liu and Anderson
showed for the first time elevated ceramide levels with consequent decrease in
sphingomyelin content of caveolae compartment of human fibroblasts in response to
proinflammatory cytokine IL-lB (116). Bilderback et al. Showed that sphingomyelin
hydrolysis takes place in caveolae-rich domains in NIH 3T3 fibroblasts in response to
nerve growth factor (N GF) treatment (207). As a second messenger, ceramide plays a role
in various biological processes (122, 208); yet, the exact mechanism by which ceramide
completes its biologic functions is not completely understood. However, important
findings in the last decade suggest that changes in cell membrane structure promoted by
ceramide accumulation are essential for its biologic function (209).

Ceramide generation within the plasma membrane significantly alters the
properties of the cell membrane Since ceramide molecules spontaneously self-aggregate
(201, 210). In membrane models, as little as five mol% ceramide is enough for

spontaneous formation of ceramide-rich membrane domains (211). Furthermore, these

33

ceramide-rich domains spontaneously self-associate to generate ceramide-rich
macrodomains with a diameter up to 5pm (201 , 210, 212). The formation and existence of
ceramide-rich membrane domains was visualized in vivo by the use of confocal
microscopy employing fluorescent-labeled anti-ceramide antibody (150, 213-215).
Changes of the biophysical properties of the membrane domains may result in selective
trapping of proteins, and in particular various receptors and signaling molecules, within
these ceramide-rich membrane platforms. Clustering of receptors may lead to an increase
in receptor density, a spatial connection of activated receptors with downstream signaling
molecules, a conformational modification of the receptor and a possible stabilization of
the receptor-ligand interaction. Thus, the release of ceramides alters the dynamics of these
rafts and may drive signal transduction processes by allowing oligomerization of specific
cell surface molecules, such as ligated receptors. Among these are the TNFa receptor
(216-217), IL-1 receptor (117), CD95 (150, 156), CD40 (178), LFA-l (218), CD5 (219),
FcyRII (213), CD20 (220), platelet-activating factor (PAF) receptor, insulin receptor,
epidermal growth factor receptor, and T cell receptor are most characterized (180).
Clustering is shown to be an important feature used by several receptors that mediate
inflammatory signaling pathways in DR, such as TNFa and IL-IB pathways (221-222).
Thus, ceramide-rich macrodomains may have a primarily role to reorganize receptors and
signaling molecules at the plasma membrane level in order to facilitate and augment

signaling processes via a particular receptor (Figure 1.4) (155).

34

snhingoml [enn ........................................

 

Sphingosine H '0 fH’ ,.
cm P-O —CH,- CH,- tr- CH3;
II ='
H :-— 9 CH3
2' ....................................
Fatty acid Phosphocholine
Qeramide Sphingomyelinases
Sphingosine

 

Figure 1.3. Ceramide generation by sphingomyelinases-mediated sphingomyelin
hydrolysis. Sphingomyelin consists of a sphingosine backbone, a fatty acid linked by an
amide bond to the sphingoid base and a phosphocholine head group. Ceramide consists of
a sphingosine backbone with the fatty acid attached to the sphingoid base. Sphingomyelin
hydrolysis by Sphingomyelinases generates ceramides.

35

Microdomains

(...)

    
 

1

 

 

 

I]

TASMase, NSMase

 

 

 

 

 

III III III III

Amplification of inflammatory signal transduction

U sphingomyelin 'ceramide Dcholesterol receptors

a cytokine

Figure 1.4. Receptor clustering and amplification of signal transduction by
ceramide-enriched macrodomains. Stimulation of cells via a receptor, for example
p55TNFa receptor, activates ASMase and/or NSMase with ceramide-release from
sphingomyelin that results in formation of ceramide-enriched macrodomains. These
membrane platforms induce further receptors clustering, re-organization of the
intracellular signalosome and amplification of the initial signal.

36

5. Retinal Vascular Endothelium

5.1. Overview

Vascular endothelium forms the inner blood-retinal barrier, a metabolically active
interface responsible for fatty acids uptake from the blood and delivery to the retina.
Moreover, retinal vasculature may also actively remodel fatty acids, thus providing the
retina with long chain highly unsaturated fatty acids such as DHA (223). Several studies
have shown that retinal microvessels are highly enriched in DHA (60). In retinal bovine
vasculature, DHA accounts for approximately 10% of the total fatty acids (60). Therefore,
DHA may have vital role in maintaining retinal capillary structure and integrity (62).
Moreover, DHA supplementation in cytokine-stimulated human retinal endothelial cells
has a pronounced anti-inflammatory effect (3 9-40).

Retinal vascular endothelial cells are the resident vasculature affected in DR.
Although diabetes may also result in damage to nonvascular retinal tissue, retinal
microangiopathy lesions represent the key pathologic feature of DR. Diabetic retinal
endothelial dysfunction includes a number of functional alterations in the vascular
endothelium, including inflammatory activation, cellular apoptosis and altered blood-
retinal barrier fimction. Many of the molecular and functional changes characteristic of
inflammation have been detected in retinas from diabetic animal models and humans and
are recognized to be an early trigger for endothelial cell activation and apoptosis and
firrther development of degenerated, non-perfused capillaries. The inflammatory
mechanisms in retinal endothelial cells appear to be mediated at the plasma membrane
level where the integrity of membrane microdomains are essential for inflammatory

Signal transduction (39).

37

5.2. Sphingomyelinases and Vascular Endothelium

Vascular endothelial cells, such as human retinal endothelial cells, represent one
of the cell type that possess lipid rafts and numerous caveolae structures with high levels
of caveolin-1 (224). Membrane domains, by virtue of their unique lipid composition,
serve as signaling platforms that regulate various cellular events (225). Sphingolipids
(Sphingomyelin, ceramide) represent an essential component of membrane microdomains
and, as previously described, sphingomyelinases-induced ceramide enrichment within
caveolae/lipid rafts would promote receptors’ clustering that is a prerequisite for the
Signalosome organization and amplification of intracellular inflammatory cascade.

Although sphingomyelinases are ubiquitously expressed in various tissues and
organs (128, 226), vascular endothelial cells are well recognized to be an important
source of ASMase (119, 135), containing as much as 20 times more ASMase than other
cell types in the body (147). Moreover, human coronary artery and human umbilical vein
endothelial cells, as well as bovine and murine aortic endothelial cells secrete copious
amounts of ASMase in response to inflammatory cytokines (119). Importantly, an
ASMase isoform has been identified in the caveolar fractions isolated from pulmonary
vascular endothelial cells (227), suggesting a major role for ASMase in local ceramide
production and amplification of cellular inflammatory transduction in pulmonary vascular
endothelium. Several studies have identified ASMase activation as a triggering
mechanism and driving force for membrane microdomain clustering in coronary

endothelial cells and consequent endothelial dysfunction (228-230).

Retinal vascular inflammation and degeneration is the hallmark microangiopathic

lesion characteristic for DR. Therefore, in this study, we first examined in vitro the role of

38

sphingomyelinases in mediating cytokine-induced inflammatory signaling in human
retinal endothelial cells; then we assessed sphingomyelinases involvement in retinal

inflammation and vascular degeneration in in vivo animal models of retinopathy.

39

6. Hypothesis and Objective of the Thesis

Sphingomyelinases, key regulatory enzymes of sphingolipid metabolism, are
important early responders in inflammatory cytokine signaling that catalyze the
hydrolysis of sphingomyelin to the proinflammatory bioactive second messenger
ceramide. SMases are critically involved in pathophysiology of various metabolic
disorders (47) and play an active role in cellular signaling (124). ASMase is shown to be
significantly upregulated in plasma (231), skeletal muscle (57) or adipose tissue (232) of
diabetic humans and animal models, whereas increased NSMase has been described in
diabetic skeletal muscle and adipose tissue (57, 232-233). However, their role in
mediating retinal inflammation and vascular degeneration in diabetic retinopathy has not
been described.

Omega 3-PUFAS and particularly DHA have the ability to modulate various
biological processes involved in retinal vascular inflammation, retinal capillary structure
and integrity alterations, and retinal neovascularization. The major (o3-PUFA DHA is
decreased in diabetic retina and associates with the inflammatory status in the diabetic
retinal tissue (61). DHA is well recognized to have anti-inflammatory properties and our
previous study demonstrated that DHA inhibits cytokine-induced retinal endothelial cells
activation through modification of caveolae membrane microdomain (39-40). The role of
DHA in sphingolipid methabolism is not known and will represent one of the goals of this
dissertation.

Various inflammatory Signaling pathways involved in the pathogenesis of diabetic
retinopathy, such as TNFa and IL-IB pathways, are initiated at the receptors located in

the caveolae/lipid rafts and ceramide enrichment in these membrane microdomains

40

promotes receptors” clustering that is a prerequisite for the intracellular inflammatory
cascade. Therefore, we hypothesize in this study that sphingomyelinases play an
important role in modulating intracellular inflammatory Signaling in retinal vascular
endothelial cells and inhibition of sphingomyelinases will render protection against
cytokine-induced endothelium activation.

Taken all together, in this dissertation we set our goal to determine the effect of
inflammatory cytokines on sphingomyelinases activation; the role of sphingomyelinases
in mediating cytokine-induced inflammation; the caveolae microdomains lipid
composition with regards to sphingomyelin and ceramide molecular species in cells
where sphingomyelinases activation and inhibition is attained and the subsequent impact
on cytokine-induced inflammatory Signaling. The studies will be conducted in primary
vascular endothelial cells isolated from human retina as a cell culture model.
Furthermore, the role of sphingomyelinases in mediation of inflammatory process
obtained in the cell culture setting will be confirmed and characterized in in vivo studies
using three complementary animal models of retinopathy: 1) streptozotocin (STZ)-
induced diabetes rat model, 2) retinal ischemia-reperfusion mouse and rat models and 3)

oxygen-induced retinopathy mouse model.

Overview of chapters
Chapter I is dedicated to a literature review of diabetic retinopathy, the role of
dyslipidemia and (0-3 PUFAS (especially DHA) in mediating retinal inflammation, the

inflammatory mechanisms in diabetic retinopathy that result in endothelial cell

41

dysfunction and death, and involvement of sphingomyelinases (sphingomyelin/ceramide
pathway) in inflammatory signal transduction.

Chapter II establishes the anti-inflammatory role of DHA by decreasing basal and
cytokine-induced activation of ASMase and NSMase, whereas inhibition and/or gene
silencing of both SMases recapitulates the DHA protective effect against cytokine-
induced inflammatory signaling in human retinal endothelial cells.

Chapter III determines the anti-inflammatory effect of regulation of caveolar
ASMase expression by DHA and the impact on caveolae/lipid membrane microdomains
sphingomyelin and ceramide molecular species that results in decreased cytokine-induced
inflammatory signaling in human retinal endothelial cells.

Chapter IV demonstrates ASMase downregulation by DHA as a protective, anti-
inflammatory effect in diabetic retinopathy that prevents retinal vessel loss in in vivo
model of type 1 diabetic rat.

Chapter V characterizes the decreased inflammatory profile and reduced retinal
vessel loss in the ASMase deficient (ASMase/j when compared to wild type (ASMase+/+)
mouse model of retinal ischemia-reperfusion, as well as the protective role of ASMase
genetic deficiency against retinal pathological neovascularization in the oxygen-induced
retinopathy mouse model.

The results of this work will add to our understanding the role of
sphingomyelin/ceramide pathway in mediating retinal vascular inflammation, vessel loss

and aberrant angiogenesis and facilitate novel therapeutic and preventive strategies for

development and progression of diabetic retinal microangiopathic lesions.

42

 

Diabetes I
/\

Hyperglycemia I Dyslipidemia I
l

i w 3 PUFAS (DHA) in the retina I

/

Pro-inflammatory state in the retina J

rCytokines (TNFa, lL-1B) If " - “>I ASMase upregulationd—DHA

\/

Retinal vascular degenerationJ

Diabetic retinopathyI

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1.5. Diagram of working hypothesis. Potential impact of decreased retinal (o3-
PUFA DHA on retinal inflammatory status, ASMase upregulation, microvascular
complications and progression of diabetic retinopathy.

43

7. References

l.

10.

ll.

12.

Fong, D.S., Aiello, L.P., Fenis, F.L., 3rd, and Klein, R. 2004. Diabetic
retinopathy. Diabetes Care 27:2540-2553.

Varma, R. 2008. From a population to patients: the Wisconsin epidemiologic
study of diabetic retinopathy. Ophthalmology 1 15: 1857-1858.

Aiello, L.P., Cavallerano, J., and Bursell, SE. 1996. Diabetic eye disease.
Endocrinol Metab Clin North Am 25:271-291.

Rein, D.B., Zhang, R, Wirth, K.E., Lee, P.P., Hoerger, T.J., McCall, N., Klein,
R., Tielsch, J.M., Vijan, S., and Saaddine, J. 2006. The economic burden of major
adult visual disorders in the United States. Arch Ophthalmol 124:1754-1760.

Saaddine, J.B., Honeycutt, A.A., Narayan, K.M., Zhang, X., Klein, R., and Boyle,
LP. 2008. Projection of diabetic retinopathy and other major eye diseases among
people with diabetes mellitus: United States, 2005-2050. Arch Ophthalmol
126:1740-1747.

Sinclair, A.J., Bayer, A.J., Girling, A.J., and Woodhouse, K.W. 2000. Older
adults, diabetes mellitus and visual acuity: a community-based case-control study.
Age Ageing 29:335-339.

Kern, TS 2007. Contributions of inflammatory processes to the development of
the early stages of diabetic retinopathy. Exp Diabetes Res 2007:95103.

Lee, R, Wang, CC, and Adamis, AP. 1998. Ocular neovascularization: an
epidemiologic review. Surv Ophthalmol 43:245-269.

Klein, R., Klein, B.E., Moss, S.E., Davis, M.D., and DeMets, D.L. 1984. The
Wisconsin epidemiologic study of diabetic retinopathy. II. Prevalence and risk of

diabetic retinopathy when age at diagnosis is less than 30 years. Arch Ophthalmol
102:520-526.

Klein, R., Klein, B.E., Moss, S.E., Davis, M.D., and DeMets, D.L. 1984. The
Wisconsin epidemiologic study of diabetic retinopathy. III. Prevalence and risk of

diabetic retinopathy when age at diagnosis is 30 or more years. Arch Ophthalmol
102:527-532.

Frank, RN. 2004. Diabetic retinopathy. N Engl J Med 350:48-58.
Ishida, S., Usui, T., Yamashiro, K., Kaji, Y., Amano, S., Ogura, Y., Hida, T.,
Oguchi, Y., Ambati, J., Miller, J.W., et al. 2003. VEGF164-mediated

inflammation is required for pathological, but not physiological, ischemia-induced
retinal neovascularization. J Exp Med 198:483-489.

44

l3.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

Klein, R., Klein, B.E., Moss, S.E., and Cruickshanks, K]. 1994. The Wisconsin
Epidemiologic Study of diabetic retinopathy. XIV. Ten-year incidence and
progression of diabetic retinopathy. Arch Ophthalmol 112:1217-1228.

Klein, R., Klein, B.E., Moss, S.E., Davis, M.D., and DeMets, D.L. 1984. The
Wisconsin epidemiologic study of diabetic retinopathy. IV. Diabetic macular
edema. Ophthalmology 91 :1464-1474.

Lyons, T.J., Jenkins, A.J., Zheng, D., Lackland, D.T., McGee, D., Garvey, WT,
and Klein, R.L. 2004. Diabetic retinopathy and serum lipoprotein subclasses in
the DCCT/EDIC cohort. Invest Ophthalmol Vis Sci 45:910-918.

Kissebah, A.H., Kohner, E.M., Lewis, B., Siddiq, Y.K., Lowy, C., and Fraser,
TR. 1975. Plasma-lipids and glucose/insulin relationship in non-insulin-requiring
diabetics with and without retinopathy. Lancet 1:1104-1 108.

Ferris, F .L., 3rd, Chew, E.Y., and Hoogwerf, 8.1. 1996. Serum lipids and diabetic
retinopathy. Early Treatment Diabetic Retinopathy Study Research Group.
Diabetes Care 19:1291-1293.

Chew, E.Y., Klein, M.L., Ferris, F.L., 3rd, Remaley, N.A., Murphy, R.P.,
Chantry, K., Hoogwerf, B.J., and Miller, D. 1996. Association of elevated serum
lipid levels with retinal hard exudate in diabetic retinopathy. Early Treatment
Diabetic Retinopathy Study (ETDRS) Report 22. Arch Ophthalmol 114:1079-
1084.

Van Eck, W.F. 1959. The effect of a low fat diet on the serum lipids in diabetes
and its significance in diabetic retinopathy. Am J Med 27:196-21 1.

Duncan, L.J., Cullen, J .F ., Ireland, J .T., Nolan, J ., Clarke, B.F., and Oliver, M.F.
1968. A three-year trial of atromid therapy in exudative diabetic retinopathy.
Diabetes 17:458-467.

Houtsmuller, A.J., Zahn, K.J., and Henkes, HE. 1980. Unsaturated fats and
progression of diabetic retinopathy. Doc Ophthalmol 48:363-371.

Kreisberg, RA. 1998. Diabetic dyslipidemia. Am J Cardiol 82:67U-73U;
discussion 85U-86U.

Coppack, S.W., Evans, R.D., Fisher, R.M., Frayn, K.N., Gibbons, G.F.,
Humphreys, S.M., Kirk, M.L., Potts, J.L., and Hockaday, TD. 1992. Adipose
tissue metabolism in obesity: lipase action in vivo before and after a mixed meal.
Metabolism 41:264-272.

45

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

Weinstock, P.H., Levak-Frank, S., Hudgins, L.C., Radner, H., Friedman, J.M.,
Zechner, R., and Breslow, J .L. 1997. Lipoprotein lipase controls fatty acid entry
into adipose tissue, but fat mass is preserved by endogenous synthesis in mice
deficient in adipose tissue lipoprotein lipase. Proc Natl Acad Sci U S A 94:10261-
10266.

Goldberg, RB. 1981. Lipid disorders in diabetes. Diabetes Care 4:561-572.

Brenner, RR. 2003. Hormonal modulation of delta6 and delta5 desaturases: case
of diabetes. Prostaglandins Leukot Essent Fatty Acids 68: 151-162.

Goldberg, U. 2001. Clinical review 124: Diabetic dyslipidemia: causes and
consequences. J Clin Endocrinol Metab 86:965-971.

Goldberg, RB, and Capuzzi, D. 2001. Lipid disorders in type 1 and type 2
diabetes. Clin Lab Med 21 :147-172, vii.

Poisson, J .P. 1989. Essential fatty acid metabolism in diabetes. Nutrition 5:263-
266.

Vessby, B. 2000. Dietary fat and insulin action in humans. Br J Nutr 83 Suppl
1:S91-96.

Bierman, EL. 1992. George Lyman Duff Memorial Lecture. Atherogenesis in
diabetes. Arterioscler T hromb 12:647-656.

Syvanne, M., and Taskinen, MR. 1997. Lipids and lipoproteins as coronary risk
factors in non-insulin-dependent diabetes mellitus. Lancet 350 Suppl 1:8120-23.

Solano, M.P., and Goldberg, RB. 2006. Management of dyslipidemia in diabetes.
Cardiol Rev 14:125-135.

Stampfer, M.J., Krauss, R.M., Ma, J., Blanche, P.J., Holl, L.G., Sacks, F .M., and
Hennekens, CH. 1996. A prospective study of triglyceride level, low-density

lipoprotein particle diameter, and risk of myocardial infarction. Jama 276:882-
888.

Perez, A., Wagner, A.M., Carreras, G., Gimenez, G., Sanchez-Quesada, J.L.,

’ Rigla, M., Gomez-Gerique, J.A., Pou, J .M., and de Leiva, A. 2000. Prevalence

and phenotypic distribution of dyslipidemia in type 1 diabetes mellitus: effect of
glycemic control. Arch Intern Med 160:2756-2762.

Kordonouri, 0., Danne, T., Hopfenmuller, W., Enders, I., Hovener, G., and
Weber, B. 1996. Lipid profiles and blood pressure: are they risk factors for the
development of early background retinopathy and incipient nephropathy in
children with insulin-dependent diabetes mellitus? Acta Paediatr 85:43-48.

46

37.

38.

39.

40.

41.

42.

43.

45.

46.

47.

48.

Calder, P.C. 2003. N-3 polyunsaturated fatty acids and inflammation: from
molecular biology to the clinic. Lipids 38:343-3 52.

Calder, P.C. 2006. n-3 polyunsaturated fatty acids, inflammation, and
inflammatory diseases. Am J Clin Nutr 83: 1 505S-1519S.

Chen, W., Jump, D.B., Esselman, W.J., and Busik, J.V. 2007. Inhibition of
cytokine signaling in human retina] endothelial cells through modification of
caveolae/lipid rafts by docosahexaenoic acid. Invest Ophthalmol Vis Sci 48:18-26.

Chen, W., Esselman, W.J., Jump, DB, and Busik, J .V. 2005. Anti-inflammatory
effect of docosahexaenoic acid on cytokine-induced adhesion molecule

expression in human retinal vascular endothelial cells. Invest Ophthalmol Vis Sci
46:4342-4347.

Simopoulos, AP. 1997. Omega-6/omega-3 fatty acid ratio and trans fatty acids in
non-insulin-dependent diabetes mellitus. Ann N Y A cad Sci 827:327-33 8.

Simopoulos, AP. 2003. Importance of the ratio of omega-6/omega-3 essential
fatty acids: evolutionary aspects. World Rev Nutr Diet 9221-22.

Simopoulos, AP. 2008. The importance of the omega-6/omega-3 fatty acid ratio
in cardiovascular disease and other chronic diseases. Exp Biol Med (Maywood)
233:674-688.

Ferrucci, L., Cherubini, A., Bandinelli, S., Bartali, B., Corsi, A., Lauretani, F.,
Martin, A., Andres-Lacueva, C., Senin, U., and Guralnik, J .M. 2006. Relationship
of plasma polyunsaturated fatty acids to circulating inflammatory markers. J Clin
Endocrinol Metab 91:439-446.

Harder, T., and Simons, K. 1997. Caveolae, DIGS, and the dynamics of
sphingolipid-cholesterol microdomains. Curr Opin Cell Biol 9:534-542.

Schroeder, R., London, E., and Brown, D. 1994. Interactions between saturated
acyl chains confer detergent resistance on lipids and g1ycosylphosphatidylinositol

(GPI)-anchored proteins: GPI-anchored proteins in liposomes and cells Show
similar behavior. Proc Natl Acad Sci U S A 91:12130-12134.

Holland, W.L., and Summers, SA. 2008. Sphingolipids, insulin resistance, and
metabolic disease: new insights from in vivo manipulation of sphingolipid
metabolism. Endocr Rev 29:381-402.

Summers, SA, and Nelson, DH. 2005. A role for sphingolipids in producing the

common features of type 2 diabetes, metabolic syndrome X, and Cushing's
syndrome. Diabetes 54: 591 -602.

47

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

Kim, M.Y., Linardic, C., Obeid, L., and Hannun, Y. 1991. Identification of
sphingomyelin turnover as an effector mechanism for the action of tumor necrosis

factor alpha and gamma-interferon. Specific role in cell differentiation. J Biol
Chem 266:484-489.

Wiegrnann, K., Schutze, S., Machleidt, T., Witte, D., and Kronke, M. 1994.
Functional dichotomy of neutral and acidic Sphingomyelinases in tumor necrosis
factor Signaling. Cell 78:1005-1015.

Chavez, J.A., Knotts, T.A., Wang, L.P., Li, G., Dobrowsky, R.T., Florant, G.L.,
and Summers, SA. 2003. A role for ceramide, but not diacylglycerol, in the

antagonism of insulin signal transduction by saturated fatty acids. J Biol Chem
278:10297-10303.

Chavez, J.A., and Summers, SA. 2003. Characterizing the effects of saturated
fatty acids on insulin signaling and ceramide and diacylglycerol accumulation in
3T3-L1 adipocytes and C2C12 myotubes. Arch Biochem Biophys 419:101-109.

Turinsky, J., O'Sullivan, D.M., and Bayly, BR 1990. 1,2-Diacylglycerol and
ceramide levels in insulin-resistant tissues of the rat in vivo. J Biol Chem
265:16880-16885.

Gorska, M., Dobrzyn, A., Zendzian-Piotrowska, M., and Gorski, J. 2004. Effect
of streptozotocin-diabetes on the functioning of the Sphingomyelin-signalling
pathway in skeletal muscles of the rat. Horm Metab Res 36: 14-21.

Adams, J.M., 2nd, Pratipanawatr, T., Berria, R., Wang, E., DeFronzo, R.A.,
Sullards, M.C., and Mandarino, LJ. 2004. Ceramide content is increased in
skeletal muscle from obese insulin-resistant humans. Diabetes 53:25-31.

Straczkowski, M., Kowalska, 1., Nikolajuk, A., Dzienis-Straczkowska, S.,
Kinalska, I., Baranowski, M., Zendzian-Piotrowska, M., Brzezinska, Z., and
Gorski, J. 2004. Relationship between insulin sensitivity and sphingomyelin
signaling pathway in human skeletal muscle. Diabetes 53: 1215-1221.

Straczkowski, M., Kowalska, 1., Baranowski, M., Nikolajuk, A., Otziomek, E.,
Zabielski, P., Adamska, A., Blachnio, A., Gorski, J., and Gorska, M. 2007.
Increased skeletal muscle ceramide level in men at risk of developing type 2
diabetes. Diabetologia 50:2366-23 73.

Neschen, S., Moore, 1., Regittnig, W., Yu, C.L., Wang, Y., Pypaert, M., Petersen,
K.F., and Shulman, G.I. 2002. Contrasting effects of fish oil and safflower oil on
hepatic peroxisomal and tissue lipid content. Am J Physiol Endocrinol Metab
282:E395-401.

48

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

Fox, T.E., Han, X., Kelly, S., Merrill, A.H., 2nd, Martin, R.E., Anderson, R.E.,
Gardner, T.W., and Kester, M. 2006. Diabetes alters sphingolipid metabolism in
the retina: a potential mechanism of cell death in diabetic retinopathy. Diabetes
55:3573-3580.

Lecomte, M., Paget, C., Ruggiero, D., Wiemsperger, N., and Lagarde, M. 1996.
Docosahexaenoic acid is a major n-3 polyunsaturated fatty acid in bovine retinal
microvessels. J Neurochem 66:2 160-2167.

Tikhonenko, M., Lydic, T.A., Wang, Y., Chen, W., Opreanu, M., Sochacki, A.,
McSorley, K.M., Renis, R.L., Kern, T., Jump, D.B., et al. 2009. Remodeling of
Retinal Fatty Acids in an Animal Model of Diabetes: a Decrease in Long Chain
Polyunsaturated Fatty Acids is Associated with a Decrease in Fatty Acid
Elongases Elov12 and Elovl4. Diabetes.

SanGiovanni, JP, and Chew, E.Y. 2005. The role of omega-3 long-chain
polyunsaturated fatty acids in health and disease of the retina. Prog Retin Eye Res
24287-138.

Meyer, A., Kirsch, H., Domergue, F., Abbadi, A., Sperling, P., Bauer, J ., Cirpus,
P., Zank, T.K., Moreau, H., Roscoe, T.J., et a1. 2004. Novel fatty acid elongases
and their use for the reconstitution of docosahexaenoic acid biosynthesis. J Lipid
Res 45: 1 899-1909.

Lagali, P.S., Liu, J., Ambasudhan, R., Kakuk, L.E., Bernstein, S.L., Seigel, G.M.,
Wong, P.W., and Ayyagari, R. 2003. Evolutionarily conserved ELOVL4 gene
expression in the vertebrate retina. Invest Ophthalmol Vis Sci 44:2841-2850.

Umeda, S., Ayyagari, R., Suzuki, M.T., Ono, F., Iwata, F., Fujiki, K., Kanai, A.,
Takada, Y., Yoshikawa, Y., Tanaka, Y., et a1. 2003. Molecular cloning of
ELOVL4 gene from cynomolgus monkey (Macaca fascicularis). Exp Anim
52:129-135.

Zhang, X.M., Yang, Z., Karan, G., Hashimoto, T., Baehr, W., Yang, X.J., and
Zhang, K. 2003. Elovl4 mRNA distribution in the developing mouse retina and
phylogenetic conservation of Elovl4 genes. Mol Vis 92301-307.

Decsi, T., Minda, H., Hermann, R., Kozari, A., Erhardt, E., Burns, 1., Molnar, S.,
and Soltesz, G. 2002. Polyunsaturated fatty acids in plasma and erythrocyte
membrane lipids of diabetic children. Prostaglandins Leukot Essent Fatty Acids
67:203-210.

Futterman, S., and Kupfer, C. 1968. The fatty acid composition of the retinal
vasculature of normal and diabetic human eyes. Invest Ophthalmol 7:105-108.

49

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

Hartong, D.T., Berson, EL, and Dryja, T.P. 2006. Retinitis pigmentosa. Lancet
368:1795-1809.

Hoffinan, DR, and Birch, D.G. 1995. Docosahexaenoic acid in red blood cells of
patients with X-linked retinitis pigmentosa. Invest Ophthalmol Vis Sci 3621009-
1018.

Ren, H., Magulike, N., Ghebremeskel, K., and Crawford, M. 2006. Primary open-
angle glaucoma patients have reduced levels of blood docosahexaenoic and
eicosapentaenoic acids. Prostaglandins Leukot Essent Fatty Acids 74: 1 57-163.

SanGiovanni, J.P., Chew, E.Y., Clemons, T.E., Davis, M.D., Ferris, F.L., 3rd,
Gensler, G.R., Kurinij, N., Lindblad, A.S., Milton, R.C., Seddon, J.M., et al.
2007. The relationship of dietary lipid intake and age-related macular
degeneration in a case-control study: AREDS Report No. 20. Arch Ophthalmol
125:671-679.

Bazan, N.G. 2006. Cell survival matters: docosahexaenoic acid signaling,
neuroprotection and photoreceptors. Trends Neurosci 29:263-271.

Connor, K.M., SanGiovanni, J.P., Lofqvist, C., Aderman, C.M., Chen, J.,
Higuchi, A., Hong, S., Pravda, E.A., Majchrzak, S., Carper, D., et al. 2007.
Increased dietary intake of omega-3-polyunsaturated fatty acids reduces
pathological retinal angiogenesis. Nat Med 13 :868-873.

Powell, ED, and Field, RA. 1964. Diabetic Retinopathy And Rheumatoid
Arthritis. Lancet 41:17-18.

Joussen, A.M., Poulaki, V., Mitsiades, N., Kirchhof, B., Koizumi, K., Dohmen,
S., and Adamis, AP. 2002. Nonsteroidal anti-inflammatory drugs prevent early
diabetic retinopathy via TNF-alpha suppression. F aseb J 16:43 8-440.

Joussen, A.M., Murata, T., Tsujikawa, A., Kirchhof, B., Bursell, S.E., and
Adamis, AP. 2001. Leukocyte-mediated endothelial cell injury and death in the
diabetic retina. Am J Pathol 158:147-152.

McLeod, D.S., Lefer, D.J., Merges, C., and Lutty, GA. 1995. Enhanced
expression of intracellular adhesion molecule-1 and P-selectin in the diabetic
human retina and choroid. Am J Pathol 147:642-653.

Joussen, A.M., Poulaki, V., Le, M.L., Koizumi, K., Esser, C., Janicki, H.,
Schraermeyer, U., Kociok, N., Fauser, S., Kirchhof, B., et al. 2004. A central role
for inflammation in the pathogenesis of diabetic retinopathy. Faseb J 18:1450-
1452.

50

80.

81.

82.

83.

84.

85.

86.

87.

88.

89.

Limb, G.A., Chignell, A.H., Green, W., LeRoy, F., and Dumonde, DC. 1996.
Distribution of TNF alpha and its reactive vascular adhesion molecules in

fibrovascular membranes of proliferative diabetic retinopathy. Br J Ophthalmol
80:168-173.

Murata, T., Nakagawa, K., Khalil, A., Ishibashi, T., lnomata, H., and Sueishi, K.
1996. The relation between expression of vascular endothelial growth factor and
breakdown of the blood-retinal barrier in diabetic rat retinas. Lab Invest 74:819-
825.

Adamis, A.P., Miller, J .W., Bemal, M.T., D'Amico, D.J., Folkman, J ., Yeo, T.K.,
and Yeo, K.T. 1994. Increased vascular endothelial growth factor levels in the

vitreous of eyes with proliferative diabetic retinopathy. Am J Ophthalmol
118:445-450.

Taylor, P.C., Williams, R.O., and Feldmann, M. 2004. Tumour necrosis factor
alpha as a therapeutic target for immune-mediated inflammatory diseases. Curr
Opin Biotechnol 15:557-563.

Eizirik, D.L., and Mandrup-Poulsen, T. 2001. A choice of death--the signal—
transduction of immune-mediated beta-cell apoptosis. Diabetologia 44:2115-
2133.

Ruan, H., and Lodish, HF. 2003. Insulin resistance in adipose tissue: direct and
indirect effects of tumor necrosis factor-alpha. Cytokine Growth Factor Rev
14:447-455.

Navarro, J .F., and Mora-Femandez, C. 2006. The role of TNF-alpha in diabetic
nephropathy: pathogenic and therapeutic implications. Cytokine Growth Factor
Rev 17:441-450.

Hawrami, K., Hitrnan, G.A., Rema, M., Snehalatha, C., Viswanathan, M.,
Rarnachandran, A., and Mohan, V. 1996. An association in non-insulin-dependent
diabetes mellitus subjects between susceptibility to retinopathy and tumor
necrosis factor polymorphism. Hum Immunol 46:49-54.

Limb, G.A., Webster, L., Soomro, H., Janikoun, S., and Shilling, J. 1999. Platelet
expression of tumour necrosis factor-alpha (TNF -a1pha), TNF receptors and
intercellular adhesion molecule-1 (ICAM-1) in patients with proliferative diabetic
retinopathy. Clin Exp Immunol 118:213-218.

Spranger, J ., Meyer-Schwickerath, R., Klein, M., Schatz, H., and Pfeiffer, A.

1995. [TNF-alpha level in the vitreous body. Increase in neovascular eye diseases
and proliferative diabetic retinopathy]. Med Klin (Munich) 90: 134-137.

51

90.

91.

92.

93.

94.

95.

96.

97.

98.

Sfikakis, P.P., Markomichelakis, N., Theodossiadis, G.P., Grigoropoulos, V.,
Katsilambros, N., and Theodossiadis, PG. 2005. Regression of sight-threatening
macular edema in type 2 diabetes following treatment with the anti-tumor
necrosis factor monoclonal antibody infliximab. Diabetes Care 28:445-447.

Aiello, L.P., Avery, R.L., Arrigg, P.G., Keyt, B.A., Jampel, H.D., Shah, S.T.,
Pasquale, L.R., Thieme, H., Iwamoto, M.A., Park, J.B., et al. 1994. Vascular
endothelial growth factor in ocular fluid of patients with diabetic retinopathy and
other retinal disorders. N Engl J Med 331 :1480-1487.

Amin, R.H., Frank, R.N., Kennedy, A., Eliott, D., Puklin, J .E., and Abrams, G.W.
1997. Vascular endothelial grth factor is present in glial cells of the retina and
optic nerve of human subjects with nonproliferative diabetic retinopathy. Invest
Ophthalmol Vis Sci 38:36-47.

Miller, J .W., Adamis, A.P., Shima, D.T., D'Amore, P.A., Moulton, R.S., O'Reilly,
M.S., Folkman, J., Dvorak, H.F., Brown, L.F., Berse, B., et a1. 1994. Vascular
endothelial grth factor/vascular permeability factor is temporally and spatially
correlated with ocular angiogenesis in a primate model. Am J Pathol 145:574-
584.

Qaum, T., Xu, Q., Joussen, A.M., Clemens, M.W., Qin, W., Miyamoto, K.,
Hassessian, H., Wiegand, S.J., Rudge, J ., Yancopoulos, G.D., et al. 2001. VEGF-

initiated blood-retinal barrier breakdown in early diabetes. Invest Ophthalmol Vis
Sci 42:2408-2413.

Aiello, L.P., Northrup, J.M., Keyt, B.A., Takagi, H., and Iwamoto, M.A. 1995.
Hypoxic regulation of vascular endothelial growth factor in retinal cells. Arch
Ophthalmol 11321538-1544.

Busik, J.V., Mohr, S., and Grant, MB. 2008. Hyperglycemia-induced reactive
oxygen species toxicity to endothelial cells is dependent on paracrine mediators.
Diabetes 57:1952-1965.

Famiglietti, E.V., Stopa, E.G., McGookin, E.D., Song, P., LeBlanc, V., and
Streeten, B.W. 2003. Immunocytochemical localization of vascular endothelial
growth factor in neurons and glial cells of human retina. Brain Res 969:195-204.

Joussen, A.M., Poulaki, V., Qin, W., Kirchhof, B., Mitsiades, N., Wiegand, S.J.,
Rudge, J., Yancopoulos, G.D., and Adamis, AP. 2002. Retinal vascular
endothelial growth factor induces intercellular adhesion molecule-1 and
endothelial nitric oxide synthase expression and initiates early diabetic retinal
leukocyte adhesion in vivo. Am J Pathol 160:501-509.

52

99.

100.

101.

102.

103.

104.

105.

106.

107.

108.

Senger, D.R., Connolly, D.T., Van de Water, L., Feder, J., and Dvorak, HF.
1990. Purification and NH2-terminal amino acid sequence of guinea pig tumor-
secreted vascular permeability factor. Cancer Res 50:1774-1778.

Ng, B.W., and Adamis, AP. 2005. Targeting angiogenesis, the underlying
disorder in neovascular age-related macular degeneration. Can J Ophthalmol
40:352-368.

Gerhardinger, C., Brown, L.F., Roy, S., Mizutani, M., Zucker, C.L., and Lorenzi,
M. 1998. Expression of vascular endothelial growth factor in the human retina
and in nonproliferative diabetic retinopathy. Am J Pathol 152:1453-1462.

Segawa, Y., Shirao, Y., Yamagishi, S., Higashide, T., Kobayashi, M., Katsuno,
K., lyobe, A., Harada, H., Sato, F., Miyata, H., et al. 1998. Upregulation of retinal
vascular endothelial growth factor mRNAs in Spontaneously diabetic rats without
ophthalmoscopic retinopathy. A possible participation of advanced glycation end
products in the development of the early phase of diabetic retinopathy.
Ophthalmic Res 30:333-339.

Aiello, L.P., Pierce, E.A., Foley, E.D., Takagi, H., Chen, H., Riddle, L., Ferrara,
N., King, G.L., and Smith, LE. 1995. Suppression of retinal neovascularization in
vivo by inhibition of vascular endothelial growth factor (VEGF) using soluble
VEGF-receptor chimeric proteins. Proc Natl Acad Sci U S A 92: 10457-10461.

Adamis, A.P., Altaweel, M., Bressler, N.M., Cunningham, E.T., Jr., Davis, M.D.,
Goldbaum, M., Gonzales, C., Guyer, D.R., Barrett, K., and Patel, M. 2006.
Changes in retinal neovascularization after pegaptanib (Macugen) therapy in
diabetic individuals. Ophthalmology 113:23-28.

Arevalo, J .F., Garcia-Amaris, R.A., Roca, J .A., Sanchez, J .G., Wu, L., Berrocal,
M.H., and Maia, M. 2007. Primary intravitreal bevacizumab for the management
of pseudophakic cystoid macular edema: pilot study of the Pan-American
Collaborative Retina Study Group. J Cataract Refract Surg 33:2098-2105.

Ng, E.W., and Adamis, AP. 2006. Anti-VEGF aptarner (pegaptanib) therapy for
ocular vascular diseases. Ann N Y Acad Sci 1082: 1 51-171.

Kowluru, RA, and Odenbach, S. 2004. Role of interleukin-lbeta in the
pathogenesis of diabetic retinopathy. Br J Ophthalmol 88: 1343-1347.

Vincent, J.A., and Mohr, S. 2007. Inhibition of caspase-l/interleukin-lbeta

signaling prevents degeneration of retinal capillaries in diabetes and galactosemia.
Diabetes 56:224-230.

53

109.

110.

111.

112.

113.

114.

115.

116.

117.

118.

119.

120.

121.

Kowluru, RA, and Odenbach, S. 2004. Role of interleukin-lbeta in the
development of retinopathy in rats: effect of antioxidants. Invest Ophthalmol Vis
Sci 45:4161-4166.

Mohr, S., Xi, X., Tang, J ., and Kern, TS. 2002. Caspase activation in retinas of
diabetic and galactosemic mice and diabetic patients. Diabetes 51:1172-1179.

Karin, M., and Lin, A. 2002. NF-kappaB at the crossroads of life and death. Nat
Immunol 3:221-227.

Rothwarf, D.M., and Karin, M. 1999. The NF-kappa B activation pathway: a
paradigm in information transfer from membrane to nucleus. Sci ST KE 1999:RE1.

Baldwin, A.S., Jr. 2001. Series introduction: the transcription factor NF-kappaB
and human disease. J Clin Invest 107:3-6.

Kowluru, R.A., Koppolu, P., Chakrabarti, S., and Chen, S. 2003. Diabetes-
induced activation of nuclear transcriptional factor in the retina, and its inhibition
by antioxidants. Free Radic Res 37:1 169-1180.

Romeo, G., Liu, W.H., Asnaghi, V., Kern, TS, and Lorenzi, M. 2002. Activation
of nuclear factor-kappaB induced by diabetes and high glucose regulates a
proapoptotic program in retinal pericytes. Diabetes 51:2241-2248.

Liu, P., and Anderson, R.G. 1995. Compartrnentalized production of ceramide at
the cell surface. J Biol Chem 270:27179-27185.

Mathias, S., Younes, A., Kan, C.C., Orlow, 1., Joseph, C., and Kolesnick, RN.
1993. Activation of the sphingomyelin Signaling pathway in intact EL4 cells and
in a cell-free system by IL-1 beta. Science 259:519-522.

Yang, Z., Costanzo, M., Golde, D.W., and Kolesnick, RN. 1993. Tumor necrosis
factor activation of the sphingomyelin pathway signals nuclear factor kappa B
translocation in intact HL-60 cells. J Biol Chem 268:20520-20523.

Marathe, S., Schissel, S.L., Yellin, M.J., Beatini, N., Mintzer, R., Williams, K.J.,
and Tabas, I. 1998. Human vascular endothelial cells are a rich and regulatable

source of secretory sphingomyelinase. Implications for early atherogenesis and
ceramide-mediated cell signaling. J Biol Chem 273:4081-4088.

Stylianou, E., O'Neill, L.A., Rawlinson, L., Edbrooke, M.R., Woo, P., and
Saklatvala, J. 1992. Interleukin 1 induces NF-kappa B through its type I but not
its type II receptor in lymphocytes. J Biol Chem 267:15836-15841.

Duan, RD. 2006. Alkaline sphingomyelinase: an old enzyme with novel
implications. Biochim Biophys Acta 1761:281-291.

54

122.

123.

124.

125.

126.

127.

128.

129.

130.

131.

132.

133.

Kolesnick, R. 2002. The therapeutic potential of modulating the
ceramide/Sphingomyelin pathway. J Clin Invest 110:3-8.

Pavoine, C., and Pecker, F. 2009. Sphingomyelinases: their regulation and roles in
cardiovascular pathophysiology. Cardiovasc Res 82: 1 75-1 83.

Marchesini, N., and Hannun, Y.A. 2004. Acid and neutral sphingomyelinases:
roles and mechanisms of regulation. Biochem Cell Biol 82:27-44.

Schneider, P.B., and Kennedy, E.P. 1967. Sphingomyelinase in normal human
spleens and in spleens from subjects with Niemann-Pick disease. J Lipid Res
8:202-209.

Lin, B., Hassler, D.F., Smith, G.K., Weaver, K., and Hannun, Y.A. 1998.
Purification and characterization of a membrane bound neutral pH optimum

magnesium-dependent and phosphatidylserine-stimulated sphingomyelinase from
rat brain. J Biol Chem 273:34472-34479.

Marchesini, N., Luberto, C., and Hannun, Y.A. 2003. Biochemical properties of
mammalian neutral Sphingomyelinase 2 and its role in sphingolipid metabolism. J
Biol Chem 278:13775-13783.

Goni, F.M., and Alonso, A. 2002. Sphingomyelinases: enzymology and
membrane activity. F EBS Lett 531:38-46.

Levade, T., Malagarie-Cazenave, S., Gouaze, V., Segui, B., Tardy, C., Betito, S.,
Andrieu-Abadie, N., and Cuvillier, O. 2002. Ceramide in apoptosis: a revisited
role. Neurochem Res 27:601-607.

Clarke, C.J., and Hannun, Y.A. 2006. Neutral sphingomyelinases and nSMase2:
bridging the gaps. Biochim Biophys Acta 1758:1893-1901.

Tani, M., and Hannun, Y.A. 2007. Neutral sphingomyelinase 2 is palmitoylated
on multiple cysteine residues. Role of palmitoylation in subcellular localization. J
Biol Chem 282: 10047-10056.

Tani, M., Ito, M., and Igarashi, Y. 2007. Ceramide/sphingosine/sphingosine 1-
phosphate metabolism on the cell surface and in the extracellular space. Cell
Signal 19:229-237.

Veldman, R.J., Maestre, N., Aduib, G.M., Medin, J .A., Salvayre, R., and Levade,
T. 2001. A neutral sphingomyelinase resides in sphingolipid-enriched
microdomains and is inhibited by the caveolin-scaffolding domain: potential
implications in tumour necrosis factor signalling. Biochem J 355:859-868.

55

134.

135.

136.

137.

138.

139.

140.

141.

142.

143.

144.

145.

Czamy, M., Liu, J ., Oh, P., and Schnitzer, J .E. 2003. Transient mechanoactivation
of neutral sphingomyelinase in caveolae to generate ceramide. J Biol Chem
278:4424-4430.

Andrieu-Abadie, N., and Levade, T. 2002. Sphingomyelin hydrolysis during
apoptosis. Biochim Biophys Acta 1585:126-134.

Pettus, B.J., Chalfant, CE, and Hannun, Y.A. 2002. Ceramide in apoptosis: an
overview and current perspectives. Biochim Biophys Acta 1585:] 14-125.

Clarke, C.J., Truong, T.G., and Hannun, Y.A. 2007. Role for neutral
sphingomyelinase-2 in tumor necrosis factor alpha—stimulated expression of
vascular cell adhesion molecule-1 (VCAM) and intercellular adhesion molecule-1
(ICAM) in lung epithelial cells: p38 MAPK is an upstream regulator of nSMase2.
J Biol Chem 282:1384-1396.

Strum, J .C., Ghosh, S., and Bell, RM. 1997. Lipid second messengers. A role in
cell grth regulation and cell cycle progression. Adv Exp Med Biol 407:421-431.

Pettus, B.J., Chalfant, CE, and Hannun, Y.A. 2004. Sphingolipids in
inflammation: roles and implications. Curr Mol Med 4:405-418.

Adam-Klages, S., Adam, D., Wiegmann, K., Struve, S., Kolanus, W., Schneider-
Mergener, J., and Kronke, M. 1996. FAN, a novel WD-repeat protein, couples the
p55 TNF-receptor to neutral sphingomyelinase. Cell 86:937-947.

Segui, B., Andrieu-Abadie, N., Adam-Klages, S., Meilhac, 0., Kreder, D., Garcia,
V., Bruno, A.P., Jaffrezou, J.P., Salvayre, R., Kronke, M., et a]. 1999. CD40
signals apoptosis through FAN-regulated activation of the sphingomyelin-
ceramide pathway. J Biol Chem 274:37251-3 7258.

Segui, B., Cuvillier, O., Adam-Klages, S., Garcia, V., Malagarie-Cazenave, S.,
Leveque, S., Caspar-Bauguil, S., Coudert, J., Salvayre, R., Kronke, M., et a].
2001. Involvement of FAN in TNF-induced apoptosis. J Clin Invest 108:143-151.

Rethy, LA. 2000. Growth regulation, acid sphingomyelinase gene and genomic
imprinting: lessons from an experiment of nature. Pathol Oncol Res 6:298-300.

Simonaro, C.M., Park, J.H., Eliyahu, E., Shtraizent, N., McGovem, M.M., and
Schuchman, EH. 2006. lrnprinting at the SMPD] locus: implications for acid
sphingomyelinase-deficient Niemann-Pick disease. Am J Hum Genet 78:865-870.

Ferlinz, K., Hurwitz, R., Moczall, H., Lansmann, S., Schuchman, E.H., and
Sandhoff, K. 1997. Functional characterization of the N-glycosylation Sites of
human acid sphingomyelinase by site-directed mutagenesis. Eur J Biochem
243:511-517.

56

146.

147.

148.

149.

150.

151.

152.

153.

154.

155.

156.

157.

Schuchman, E.H., Levran, 0., Pereira, L.V., and Desnick, R.J. 1992. Structural
organization and complete nucleotide sequence of the gene encoding human acid
sphingomyelinase (SMPDl). Genomics 12:197-205.

Garcia-Barros, M., Paris, F., Cordon-Cardo, C., Lyden, D., Rafii, S., Haimovitz-
Friedman, A., Fuks, Z., and Kolesnick, R. 2003. Tumor response to radiotherapy
regulated by endothelial cell apoptosis. Science 300:1155-1159.

Jensen, J .M., Schutze, S., For], M., Kronke, M., and Proksch, E. 1999. Roles for
tumor necrosis factor receptor p55 and Sphingomyelinase in repairing the
cutaneous permeability barrier. J Clin Invest 104: 1761-1770.

Chavakis, E., and Dimmeler, S. 2002. Regulation of endothelial cell survival and
apoptosis during angiogenesis. Arterioscler T hromb Vasc Biol 22:887-893.

Grassme, H., Schwarz, H., and Gulbins, E. 2001. Molecular mechanisms of
ceramide-mediated CD95 clustering. Biochem Biophys Res Commun 284:1016-
1030.

Schissel, S.L., Keesler, G.A., Schuchman, E.H., Williams, K.J., and Tabas, I.
1998. The cellular trafficking and zinc dependence of secretory and lysosomal
sphingomyelinase, two products of the acid sphingomyelinase gene. J Biol Chem
273:18250-18259.

Schissel, S.L., Schuchman, E.H., Williams, K.J., and Tabas, 1. 1996. Zn2+-
stimulated sphingomyelinase is secreted by many cell types and is a product of
the acid sphingomyelinase gene. J Biol Chem 271:18431-18436.

Dobrowsky, RT. 2000. Sphingolipid signalling domains floating on rafts or
buried in caves? Cell Signal 12:81-90.

Gulbins, E. 2003. Regulation of death receptor signaling and apoptosis by
ceramide. Pharmacol Res 47:393-399.

Grassme, H., Riethmuller, J., and Gulbins,. E. 2007. Biological aspects of
ceramide-enriched membrane domains. Prog Lipid Res 46:161-170.

Grassme, H., Jekle, A., Riehle, A., Schwarz, H., Berger, J., Sandhoff, K.,

Kolesnick, R., and Gulbins, E. 2001. CD95 Signaling via ceramide-rich membrane
rafts. J Biol Chem 276:20589-20596.

Zeidan, Y.H., and Hannun, Y.A. 2007. Activation of acid sphingomyelinase by
protein kinase Cdelta-mediated phosphorylation. J Biol Chem 282:] 1549-11561.

57

158.

159.

160.

161.

162.

163.

164.

165.

166.

167.

168.

169.

Callahan, J.W., Jones, C.S., Davidson, D.J., and Shankaran, P. 1983. The active
site of lysosomal sphingomyelinase: evidence for the involvement of hydrophobic
and ionic groups. J Neurosci Res 10:151-163.

Schissel, S.L., Jiang, X., Tweedie-Hardman, J ., Jeong, T., Camejo, E.H., Najib, J .,
Rapp, J.H., Williams, K.J., and Tabas, I. 1998. Secretory sphingomyelinase, a
product of the acid sphingomyelinase gene, can hydrolyze atherogenic

lipoproteins at neutral pH. Implications for atherosclerotic lesion development. J
Biol Chem 273:2738-2746.

Tabas, I. 1999. Secretory sphingomyelinase. Chem Phys Lipids 102:123-130.
Schutze, S., Machleidt, T., and Kronke, M. 1992. Mechanisms of tumor necrosis
factor action. Semin Oncol 19:16-24.

Schutze, S., Machleidt, T., and Kronke, M. 1994. The role of diacylglycerol and
ceramide in tumor necrosis factor and interleukin-1 signal transduction. J Leukoc
Biol 56:533-541.

Cifone, M.G., Roncaioli, P., De Maria, R., Camarda, G., Santoni, A., Ruberti, G.,
and Testi, R. 1995. Multiple pathways originate at the FaS/APO-l (CD95)
receptor: sequential involvement of phosphatidylcholine-specific phospholipase C
and acidic sphingomyelinase in the propagation of the apoptotic signal. Embo J
14:5859-5868.

Gomez-Munoz, A., Kong, J ., Salh, B., and Steinbrecher, U.P. 2003. Sphingosine-
l-phosphate inhibits acid sphingomyelinase and blocks apoptosis in macrophages.
FEBS Lett 539:56-60.

Schwandner, R., Wiegrnann, K., Bernardo, K., Kreder, D., and Kronke, M. 1998.
TNF receptor death domain-associated proteins T RADD and FADD signal
activation of acid sphingomyelinase. J Biol Chem 273:5916-5922.

Bilderback, T.R., Gazula, V.R., and Dobrowsky, RT. 2001. Phosphoinositide 3-
kinase regulates crosstalk between Trk A tyrosine kinase and p75(NTR)-
dependent sphingolipid signaling pathways. J Neurochem 76: 1 540-1551.

Dbaibo, G.S., Pushkareva, M.Y., Rachid, R.A., Alter, N., Smyth, M.J., Obeid,
L.M., and Hannun, Y.A. 1998. p53-dependent ceramide response to genotoxic
stress. J Clin Invest 1022329-339.

Quintem, LE, and Sandhoff, K. 1991. Human acid sphingomyelinase from
human urine. Methods Enzymol 197:536-540.

De Nadai, C., Sestili, P., Cantoni, 0., Lievremont, J.P., Sciorati, C., Barsacchi, R.,
Moncada, S., Meldolesi, J., and Clementi, E. 2000. Nitric oxide inhibits tumor

58

170.

171.

172.

173.

174.

175.

176.

177.

178.

179.

180.

necrosis factor-alpha-induced apoptosis by reducing the generation of ceramide.
Proc Natl Acad Sci U S A 97:5480-5485.

Santana, P., Pena, L.A., Haimovitz-Friedman, A., Martin, S., Green, D.,
McLoughlin, M., Cordon-Cardo, C., Schuchman, E.H., Fuks, Z., and Kolesnick,
R. 1996. Acid sphingomyelinase-deficient human lyrnphoblasts and mice are
defective in radiation-induced apoptosis. Cell 86: 1 89-199.

Smith, E.L., and Schuchman, EH. 2008. Acid sphingomyelinase overexpression
enhances the antineoplastic effects of irradiation in vitro and in vivo. Mol T her
16:1565-1571.

Grammatikos, G., Teichgraber, V., Carpinteiro, A., Trarbach, T., Weller, M.,
Hengge, U.R., and Gulbins, E. 2007. Overexpression of acid sphingomyelinase
sensitizes glioma cells to chemotherapy. Antioxid Redox Signal 9: 1449-1456.

Llacuna, L., Mari, M., Garcia-Ruiz, C., Fernandez-Checa, J .C., and Morales, A.
2006. Critical role of acidic sphingomyelinase in murine hepatic ischemia-
reperfusion injury. Hepatology 44:561-572.

Zumbansen, M., and Stoffe], W. 1997 . Tumor necrosis factor alpha activates NF-

kappaB in acid sphingomyelinase-deficient mouse embryonic fibroblasts. J Biol
Chem 272:10904-10909.

Lin, T., Genestier, L., Pinkoski, M.J., Castro, A., Nicholas, 8., Mogil, R., Paris,
F., Fuks, Z., Schuchman, E.H., Kolesnick, R.N., et a]. 2000. Role of acidic
Sphingomyelinase in Fas/CD95-mediated cell death. J Biol Chem 275:8657-8663.

Schuchman, EH. 2009. The pathogenesis and treatment of acid
sphingomyelinase-deficient Niemann-Pick disease. Int J Clin Pharmacol Ther 47

Suppl 1:S48-57.

Schuchman, EH. 2009. Acid sphingomyelinase, cell membranes and human
disease: Lessons from Niemann-Pick disease. FEBS Lett.

Grassme, H., Jendrossek, V., Bock, J., Riehle, A., and Gulbins, E. 2002.
Ceramide-rich membrane rafts mediate CD40 clustering. J Immunol 168:298-307.

Gulbins, E., and Kolesnick, R. 2003. Raft ceramide in molecular medicine.
Oncogene 22:7070-7077.

Gulbins, E., and Li, PL. 2006. Physiological and pathophysiological aspects of
ceramide. Am J Physiol Regul Integr Comp Physiol 290:R1 1-26.

59

181.

182.

183.

184.

185.

186.

187.

188.

189.

190.

191.

192.

Brown, DA, and London, E. 1997. Structure of detergent-resistant membrane
domains: does phase separation occur in biological membranes? Biochem Biophys
Res Commun 240:1-7.

Hooper, NM. 1998. Membrane biology: do glycolipid microdomains really exist?
Curr Biol 82R114-116.

Schroeder, R.J., Ahmed, S.N., Zhu, Y., London, E., and Brown, DA. 1998.
Cholesterol and sphingolipid enhance the Triton X-100 insolubility of
glycosylphosphatidylinositol-anchored proteins by promoting the formation of
detergent-insoluble ordered membrane domains. J Biol Chem 273:] 150-1157.

Palade, GE. 1953. Fine structure of blood capillaries. J Appl Physiol 2421424—
1436.

Schnitzer, J.E., McIntosh, D.P., Dvorak, A.M., Liu, J., and Oh, P. 1995.
Separation of caveolae from associated microdomains of GPI-anchored proteins.
Science 269: 1435-1439.

Frank, P.G., Woodman, S.E., Park, D.S., and Lisanti, M.F. 2003. Caveolin,
caveolae, and endothelial cell function. Arterioscler Thromb Vasc Biol 23:1161-
1168.

Miyawaki-Shimizu, K., Predescu, D., Shimizu, J ., Broman, M., Predescu, S., and
Malik, AB. 2006. siRNA-induced caveolin-1 knockdown in mice increases lung

vascular permeability via the junctional pathway. Am J Physiol Lung Cell Mol
Physiol 290:L405-413.

Li, 8., Song, KS, and Lisanti, MP. 1996. Expression and characterization of
recombinant caveolin. Purification by polyhistidine tagging and cholesterol-
dependent incorporation into defined lipid membranes. J Biol Chem 271:568-573.

Murata, M., Peranen, J., Schreiner, R., Wieland, F., Kurzchalia, T.V., and Simons,
K. 1995. VIP21/caveolin is a cholesterol-binding protein. Proc Natl Acad Sci U S
A 92:10339-10343.

Anderson, R.G. 1993. Caveolae: where incoming and outgoing messengers meet.
Proc Natl Acad Sci U S A 90:10909-10913.

Lisanti, M.P., Tang, Z., Scherer, P.E., Kubler, E., Koleske, A.J., and Sargiacomo,
M. 1995. Caveolae, transmembrane signalling and cellular transformation. M01

Membr Biol 12:121-124.

Anderson, R.G., and Jacobson, K. 2002. A role for lipid shells in targeting
proteins to caveolae, rafts, and other lipid domains. Science 296:1821-1825.

6O

193.

194.

195.

196.

197.

198.

199.

200.

201.

202.

203.

204.

205.

Razani, B., and Lisanti, MP. 2001. Caveolins and caveolae: molecular and
fimctional relationships. Exp Cell Res 271236-44.

Edidin, M. 2003. The state of lipid rafts: from model membranes to cells. Annu
Rev Biophys Biomol Struct 32:257-283.

Sargiacomo, M., Sudol, M., Tang, Z., and Lisanti, M.F. 1993. Signal transducing
molecules and g1ycosyl-phosphatidylinositol-linked proteins form a caveolin-rich
insoluble complex in MDCK cells. J Cell Biol 122:789-807.

Smart, B.J., Graf, G.A., McNiven, M.A., Sessa, W.C., Engelman, J.A., Scherer,
P.B., Okarnoto, T., and Lisanti, M.F. 1999. Caveolins, liquid-ordered domains,
and signal transduction. Mol Cell Biol 19:7289-7304.

Smart, E.J., Ying, Y.S., Mineo, C., and Anderson, R.G. 1995. A detergent-free
method for purifying caveolae membrane from tissue culture cells. Proc Natl
Acad Sci USA 92:10104-10108.

Fiedler, K., Parton, R.G., Kellner, R., Etzold, T., and Simons, K. 1994. VIP36, a
novel component of glycolipid rafts and exocytic carrier vesicles in epithelial
cells. EmboJ 13:1729-1740.

Harman, L.A., Lisanti, M.P., Rodriguez-Boulan, E., and Edidin, M. 1993.
Correctly sorted molecules of a GPI-anchored protein are clustered and immobile
when they arrive at the apical surface of MDCK cells. J Cell Biol 120:353-358.

Zhang, F ., Crise, B., Su, 3., Hon, Y., Rose, J .K., Bothwell, A., and Jacobson, K.
1991. Lateral diffusion of membrane-spanning and g1ycosylphoSphatidylinositol-
linked proteins: toward establishing rules governing the lateral mobility of
membrane proteins. J Cell Biol 115:75-84.

Kolesnick, R.N., Goni, F.M., and Alonso, A. 2000. Compartrnentalization of
ceramide signaling: physical foundations and biological effects. J Cell Physiol
184:285-300.

Simons, K., and Ikonen, E. 1997. Functional rafts in cell membranes. Nature
387:569-572.

Mathias, S., Pena, LA, and Kolesnick, RN. 1998. Signal transduction of stress
via ceramide. Biochem J 335 ( Pt 3):465-480.

Ballou, L.R., Laulederkind, S.J., Rosloniec, B.F., and Raghow, R. 1996. Ceramide
signalling and the immune response. Biochim Biophys Acta 1301:273-287.

Hannun, Y.A. 1996. Functions of ceramide in coordinating cellular responses to
stress. Science 274: 1 855-1 859.

61

206.

207.

208.

209.

210.

211.

212.

213.

214.

215.

216.

217.

Pena, L.A., Fuks, Z., and Kolesnick, R. 1997. Stress-induced apoptosis and the
sphingomyelin pathway. Biochem Pharmacol 53 :61 5-621.

Bilderback, T.R., Gazula, V.R., Lisanti, M.P., and Dobrowsky, RT. 1999.
Caveolin interacts with Trk A and p75(NTR) and regulates neurotrophin signaling
pathways. J Biol Chem 274:257-263.

Kolesnick, R., and Hannun, Y.A. 1999. Ceramide and apoptosis. Trends Biochem
Sci 24:224-225; author reply 227.

Stancevic, B., and Kolesnick, R. 2010. Ceramide-rich platforms in
transmembrane Signaling. F EBS Lett.

Holopainen, J .M., Subrarnanian, M., and Kinnunen, PK. 1998. Sphingomyelinase
induces lipid microdomain formation in a fluid
phosphatidylcholine/sphingomyelin membrane. Biochemistry 37: 1 7562-17570.

Veiga, M.P., Arrondo, J.L., Goni, F.M., and Alonso, A. 1999. Cerarnides in
phospholipid membranes: effects on bilayer stability and transition to nonlamellar
phases. Biophys J 76:342-350.

Nurminen, T.A., Holopainen, J.M., Zhao, H., and Kinnunen, PK. 2002.
Observation of topical catalysis by sphingomyelinase coupled to microspheres. J
Am Chem Soc 124:12129-12134.

Abdel Shakor, A.B., Kwiatkowska, K., and Sobota, A. 2004. Cell surface
ceramide generation precedes and controls F cgammaRII clustering and
phosphorylation in rafts. J Biol Chem 279:36778-36787.

Dumitru, CA, and Gulbins, E. 2006. TRAIL activates acid sphingomyelinase via

a redox mechanism and releases ceramide to trigger apoptosis. Oncogene
25:5612-5625.

Grassme, H., Bock, J ., Kun, J ., and Gulbins, E. 2002. Clustering of CD40 ligand
is required to form a functional contact with CD40. J Biol Chem 277:30289-
30299.

Natoli, G., Costanzo, A., Guido, F., Moretti, F., and Levrero, M. 1998. Apoptotic,
non-apoptotic, and anti-apoptotic pathways of tumor necrosis factor signalling.
Biochem Pharmacol 56:915-920.

Schutze, S., Potthoff, K., Machleidt, T., Berkovic, D., Wiegrnann, K., and

Kronke, M. 1992. TNF activates NF-kappa B by phosphatidylcholine-specific
phospholipase C-induced "acidic" sphingomyelin breakdown. Cell 71:765-776.

62

218.

219.

220.

221.

222.

223.

224.

225.

226.

227.

228.

Rosenman, S.J., Ganji, A.A., Tedder, T.P., and Gallatin, W.M. 1993. Syn-capping
of human T lymphocyte adhesion/activation molecules and their redistribution
during interaction with endothelial cells. J Leukoc Biol 53:1-10.

Simarro, M., Calvo, J ., Vila, J .M., Places, L., Padilla, O., Alberola-Ila, J ., Vives,
J ., and Lozano, F. 1999. Signaling through CD5 involves acidic

Sphingomyelinase, protein kinase C-zeta, mitogen-activated protein kinase kinase,
and c-Jun NHZ-termina] kinase. J Immunol 162:5149-5155.

Bezombes, C., Grazide, S., Garret, C., Fabre, C., Quillet-Mary, A., Muller, S.,
Jaffrezou, J .P., and Laurent, G. 2004. Rituximab antiproliferative effect in B-
lymphoma cells is associated with acid-Sphingomyelinase activation in raft
microdomains. Blood 104:] 166-1 173.

K0, Y.G., Lee, J.S., Kang, Y.S., Ahn, J.H., and Seo, J.S. 1999. TNF-alpha-
mediated apoptosis is initiated in caveolae-like domains. J Immunol 162:7217-
7223.

Martin, M.U., and Wesche, H. 2002. Summary and comparison of the signaling
mechanisms of the Toll/interleukin-l receptor family. Biochim Biophys Acta
1592:265-280.

Delton-Vandenbroucke, I., Grammas, P., and Anderson, RE. 1997.
Polyunsaturated fatty acid metabolism in retina] and cerebral microvascular
endothelial cells. J Lipid Res 38:147-159.

Lisanti, M.P., Scherer, P.B., Vidugiriene, J ., Tang, Z., Hermanowski-Vosatka, A.,
Tu, Y.H., Cook, RR, and Sargiacomo, M. 1994. Characterization of caveolin-
rich membrane domains isolated fi'om an endothelial-rich source: implications for
human disease. J Cell Biol 126:1] 1-126.

Ma, D.W., Seo, J., Davidson, L.A., Callaway, E.S., Fan, Y.Y., Lupton, J.R., and
Chapkin, RS. 2004. n-3 PUFA alter caveolae lipid composition and resident
protein localization in mouse colon. Faseb J 18: 1 040-1042.

Stoffel, W., Jenke, B., Block, B., Zumbansen, M., and Koebke, J. 2005. Neutral
Sphingomyelinase 2 (smpd3) in the control of postnatal grth and development.
Proc Natl Acad Sci U S A 10214554-4559.

Yang, Y., Yin, J ., Baumgartner, W., Samapati, R., Solymosi, E.A., Reppien, E.,
Kuebler, W.M., and Uhlig, S. 2009. Platelet-activating factor reduces endothelial
NO production - role of acid sphingomyelinase. Eur Respir J.

Jin, S., Zhang, Y., Yi, F., and Li, PL. 2008. Critical role of lipid raft redox

signaling platforms in endostatin-induced coronary endothelial dysfunction.
Arterioscler T hromb Vasc Biol 28:485-490.

63

229.

230.

231.

232.

233.

Bao, J.X., Xia, M., Poklis, J.L., Han, W.Q., Brimson, C., and Li, PL. 2010.
Triggering role of acid sphingomyelinase in endothelial lysosome-membrane
firsion and dysfunction in coronary arteries. Am J Physiol Heart Circ Physiol
298:H992-H1002.

Zhang, A.Y., Yi, F., Jin, S., Xia, M., Chen, Q.Z., Gulbins, E., and Li, PL. 2007.
Acid sphingomyelinase and its redox amplification in formation of lipid raft redox
signaling platforms in endothelial cells. Antioxid Redox Signal 9:817-828.

Gorska, M., Baranczuk, E., and Dobrzyn, A. 2003. Secretory Zn2+-dependent
sphingomyelinase activity in the serum of patients with type 2 diabetes is
elevated. Harm Metab Res 35:506-507.

Samad, F., Hester, K.D., Yang, Q, Hannun, Y.A., and Bielawski, J. 2006. Altered
adipose and plasma sphingolipid metabolism in obesity: a potential mechanism

for cardiovascular and metabolic risk. Diabetes 55:2579-2587.

Summers, SA. 2006. Ceramides in insulin resistance and lipotoxicity. Prog Lipid
Res 45:42-72.

64

Chapter II
Docosahexaenoic Acid Inhibits Cytokine Signaling by Downregulating

Sphingomyelinases in Human Retinal Endothelial Cells

1. Abstract

The anti-inflammatory mechanism of docosahexaenoic acid (DHA) is well
accepted; however, the molecular mechanism(s) of DHA action are still not well defined.
We have previously demonstrated that DHA modifies the caveolar membrane
microdomain and inhibits cytokine-induced inflammation in human retina] endothelial
cells (HREC), the resident vasculature affected by diabetic retinopathy. Sphingolipids
represent a major component of plasma membrane microdomains and ceramide-enriched
microdomains appear to be a prerequisite for inflammatory cytokines signaling. Acid and
neutral sphingomyelinase (ASMase and NSMase) are key regulatory enzymes of
sphingolipid metabolism, promoting sphingomyelin hydrolysis to pro-inflammatory
ceramide. In this study, we propose to test the hypothesis that DHA inhibits cytokine-

induced inflammatory Signaling in HREC by downregulating sphingomyelinases.

Primary cultures of HREC were maintained in the presence or absence of DHA.
ASMase and NSMase activity was determined by sphingomyelinase assay. The
expression of ASMase, NSMase, ICAM-1 and VCAM-l was assessed by quantitative
PCR and Western blotting. Gene silencing of ASMase and NSMase was obtained by
siRNA treatment. Inflammatory cytokines TNFa and IL-IB induced cellular adhesion

molecules (CAMS) expression and rapid increase in ASMase and NSMase activity in

65

HREC. Growth of HREC in media supplemented with DHA decreased basal and
cytokine-induced ASMase and NSMase expression and activity, as well as upregulation
of CAMS expression. Anti-inflammatory effects of DHA on cytokine-induced CAMS
expression were mimicked by inhibition/gene silencing of ASMase and NSMase.
Inhibition of sphingomyelinases activity was mediated by DHA-induced cholesterol
displacement from caveolae/lipid microdomains. Sphingomyelinases pathway rather than

ceramide de nova synthesis pathway was important for inflammatory Si gnaling in HREC.

This study provides a novel potential mechanism for anti-inflammatory effect of
DHA in HREC. DHA downregulates basal and cytokine-induced ASMase and NSMase
activity and expression level in HREC and inhibition of sphingomyelinases in endothelial

cells prevents cytokine-induced inflammatory response.

66

2. Introduction

The retina has a unique fatty acid profile, with the highest level of polyunsaturated
fatty acids (PUFAS) in the body, especially DHA (1). Deficiency in DHA has been
documented to be associated with a number of retinal degenerative diseases, including
retinitis pigmentosa, retinopathy of prematurity and age-related macular degeneration
(reviewed in (2)). DHA is also Significantly decreased in plasma of diabetic children (3),
as well as in the human retina of diabetic eyes (4). Recent data showed a beneficial effect
of dietary DHA in reducing pathological retinal angiogenesis and thus preventing the
development of oxygen-induced retinopathy (5). Yet, the DHA protective mechanism in

retinopathy remains poorly understood.

The aim of our study was to investigate the anti-inflammatory mechanism of DHA
in human retina] endothelial cells (HREC), the target tissue affected by diabetic
retinopathy. Very early stage diabetic retinopathy is hypothesized to represent a low-
grade chronic inflammatory disease that involves leukocyte adhesion to the retinal
vasculature, a process mediated by adhesion molecules expressed on the endothelial cell
surface, especially intercellular adhesion molecule-1 (ICAM-1) and vascular cell
adhesion molecule-1 (VCAM-l) (6-8). Pro-inflammatory cytokines, including TNFa and
IL-lB are increased in diabetic eyes (9-11) and induce upregulation of adhesion molecule
expression (12). Our group has previously shown that DHA inhibits cytokine-induced
cellular adhesion molecule expression in HREC through cholesterol displacement from

caveolae/lipid rafts membrane microdomains (12).

HREC contain lipid rafts and Specific plasma membrane microdomains known as

caveolae that are considered to have an important role in regulating vascular permeability

67

(13), lipid trafficking, cholesterol homeostasis (14-15) and signal transduction (16-17).
The caveolae are stabilized by membrane spanning proteins, the caveolins, a family of
palmitoylated hairpin-like membrane proteins (18-21). Caveolin-1, the primary structural
protein constituent of caveolae in endothelial cells, is a biochemical marker for caveolae

identification (22).

Caveolae/lipid rafts are compositionally and functionally specific domains which
are not static but are rather in equilibrium with the rest of membrane lipids; they are
lateral assemblies of cholesterol, sphingolipids (sphingomyelin, ceramide) and
glycerophospholipids (23). Sphingolipids play essential roles in membrane microdomain
structure and are now known to also act as messengers in signaling pathways mediating
inflammation, apoptosis, cell differentiation and proliferation (24). Ceramides can be
generated at the membrane level by catabolism of sphingomyelin by either neutral or acid

sphingomyelinases or by de nova synthesis in the endoplasmic reticulum.

NSMase and ASMase are rapidly activated by diverse stress stimuli and promote
hydrolysis of sphingomyelin into ceramide and phosphorylcholine (25-27). NSMase is
considered one of the major candidates for mediating stress-induced production of
ceramide; it is a membrane-bound protein and palmitoylated on five cysteine residues via
thioester bonds (28-30). ASMase was initially described as a lysosomal enzyme because
of its optimal pH at 4.5—5.0. Nevertheless, an ASMase isofonn was recently Shown to be
enclosed into secretory vesicles in close proximity to the plasma membrane and to be
secreted into the extracellular Space upon cell stimulation (31-33). Liu and Anderson
described the plasma membrane form of ASMase in caveolae and Showed that the pro-

inflammatory cytokine IL-lB may induce ASMase activation in this compartment (27).

68

Lipid analysis demonstrated that approximately 70% of total cellular
sphingomyelin is found in membrane microdomains (20, 34-37). Enrichment of
sphingomyelin content in membrane microdomains makes them potential substrate pools
for cellular sphingomyelinases to produce a high local concentration of ceramide;
ceramide-enriched membrane microdomains have the ability to self-associate to form
ceramide rich macrodomains (38-41) that mediate receptor clustering and downstream
signaling (42-43). Clustering is shown to be an important feature used by several

receptors that mediate inflammatory signaling pathways in diabetic retinopathy, such as

the TNFa and IL-IB pathways (44-45).

In this study, we tested whether the anti-inflammatory mechanism of DHA is
mediated through downregulation of ASMase and NSMase, as manifested by decreased

cytokine signaling in human retina] endothelial cells.

69

3. Materials and methods

3.1. Reagents and Supplies

DMEM and F12 culture medium, antibiotics, fetal bovine serum and trypsin were
obtained from Invitrogen (Carlsbad, CA). Amplex Red Sphingomyelinase Assay Kit,
Amplex Red Cholesterol Assay Kit, NuPAGE Novex 10% Bis- Tris gels, Platinum
SYBR Green qPCR SuperMix-UDG w/ROX were purchased from Invitrogen (Carlsbad,
CA). GW4869 was purchased fiom Calbiochem (San Diego, CA). Desiprarnine, methyl-
cyclodextrine and commonly used chemicals and reagents were purchased from Sigma
(St. Louis, MO). Caveolin-1 and LAMP] antibodies (BD Bioscience, San Jose, CA),
ICAM-l, VCAM-l and NSMase antibodies (Santa Cruz Biotechnology, Santa Cruz, CA)
were used. ASMase antibody was a generous gift from Dr. Richard Kolesnick. TNFa and
IL-IB were from R&Dsystems (Minneapolis, MN). ON-TARGET plus SMART poo]
SMPDl (ASMase), ON-TARGET plus SMART pool SMPD2 (NSMase) and ON-
TARGET plus siCONTROL (non-targeting pool, against luciferase) were purchased fi'om
Dharmacon (Chicago, IL). For lipid extraction and mass spectrometry, all solvents used
were HPLC grade. Methanol (MeOH), and water were purchased from IT Baker
(Phillipsburg, NJ). Ammonium hydroxide (NH4OH) and chloroform (CHC13) were from
EMD Chemicals (Gibbstown, NJ). Isopropanol was from Fisher Scientific (Pittsburgh,

PA). Lipid standards were obtained fiom Avanti Polar Lipids (Alabaster, AL).

70

3.2. Cell Culture

Primary cultures of HREC were prepared from tissue (National Disease Research
Interchange, Philadelphia, PA) cultured as previously described (46). Passages 1 to 5
were used in the experiments. For experimental treatment, cells were transferred to

serum-free medium for 14 to 24 hours before stimulatory agents were added.

3.3. Fatty Acid Treatment

Fatty acid stocks were prepared by dissolving fatty acids (NuCheck Prep, Inc., Elysian,
MN) in ethanol to a final concentration of 100 mM fatty acid, as described previously (12,
47). The fatty acid stock solutions were diluted to 50 to 100 uM in serum-free medium
containing 10 to 20 uM charcoal-treated, solvent-extracted, fatty acid—free bovine serum
albumin (BSA; Serologica Inc., Norcross, GA) as a fatty acid carrier. The fatty
acid/albumin molar ratio was maintained at 5:1 (48). The final concentration of ethanol in
the media was less than 0.1%. Cells were incubated for the times indicated in Results.
Equivalent amounts of BSA and ethanol were added to control plates. Linoleic acid, a (06
polyunsaturated fatty acid, was used as a lipid control for the following considerations.
The most abundant polyunsaturated fatty acid classes in the retina microvessels are 033
(especially DHA) and (b6 (linoleic acid, arachidonic acid) fatty acids (1). In this study we
wanted to compare the effects of the principal 033 PUFA (DHA) with a (06 PUFA

(linoleic acid).

71

3.4. Sphingomyelinase Assay

Fatty acids (DHA and linoleic acid) were added to HREC in serum-free medium to a
concentration of 100 uM fatty acid with 20 uM charcoal-treated, solvent extracted, fatty-
acid-free bovine serum albumin (BSA) as a fatty acid carrier. Cells were incubated with
fatty acids for specified times at 37°C; equivalent amounts of BSA and ethanol were
added to the control plates. Then cells were lysed in the acid lysis buffer (50 mM Sodium
Acetate, pH=5; 1% TritonX-l 00; 1 mM EDTA) or neutral lysis buffer (20 mM Tris-HCl,
pH=7.5; 1% TritonX-IOO; 1 mM EDTA) with fieshly added protease inhibitor cocktail
(Sigma, St. Louis, MO). For the sphingomyelinase assay in a cell-fiee system, 0.5mU of
bacterial sphingomyelinase was incubated with 5-80 uM fatty acids (DHA and linoleic
acid), while equivalent amounts of ethanol were added to the control wells.
Sphingomyelinase activity was measured using the Amplex Red Sphingomyelinase

Assay Kit (Molecular Probes, Eugene, OR) as described in the manufacturer’s protocol.

3.5. Methyl-Il-cyclodextrin (MCD) and Cholesterol Treatment

To deplete cholesterol from the membranes, cells were treated with 8 mM MCD for 30
minutes. To replenish cholesterol, cells were treated with 25 uM water-soluble
cholesterol (25 uM cholesterol complexed with 250 uM MCD; Sigma) for 30 minutes.
After MCD or water-soluble cholesterol treatment, the cells were washed twice with PBS
containing 40 uM BSA before membrane microdomain isolation, and cell lysate and

plasma membrane preparation for the sphingomyelinase assay.

72

3.6. Cholesterol Measurement

Cholesterol levels of membrane microdomains isolated from HREC with or without
MCD treatment (previously described) were measured by the Amplex Red Cholesterol

Assay Kit (Molecular Probes, Eugene, OR) as described in the manufacturer’s protocol.

3.7. Western Blot Analysis

Cells were lysed in the lysis buffer (50 mM HEPES, pH 7.5, 150 mM NaCl, 1.5 mM
MgC12, 1 mM EGTA, 1% Triton X-100 and 10% glycerol) with fieshly added protease
inhibitor cocktail (Sigma) and phosphatase inhibitors (1 mM Na3VO4, 100 uM
glycerophosphate, 10 mM NaF, lmM Na4PPi). Protein concentration was measured by
Qubit fluorometer (Invitrogen, Eugene, OR) as described in the manufacturer protocol.
Proteins were resolved by NuPAGE on Novex 10% Bis-Tris gels, transferred to
nitrocellulose membrane, and then immunobloted using appropriate primary antibodies
followed by secondary horseradish peroxidase-conjugated antibody (Bio-Rad, Hercules,
CA) or lRDye infrared secondary antibodies (Invitrogen, Molecular probes, Eugene,
OR). Irnmunoreactive bands were visualized by enhanced chemiluminescence (ECL kit;
Amersham Pharrnacia Biotech, Piscataway, NJ) or the Odyssey program. Blots were

quantified by scanning densitometry (lmageJ software).

73

3.8. Gene Silencing of ASMase and NSMase

For silencing ASMase and NSMase expression, cultured HREC were detached with
trypsin, centrifuged at 100 g for 5 minutes and resuspended in electroporation solution
(Amaxa, Gaithersburg, MD) to a final concentration of 4-5x105 cells/ 100 pL. The 100 uL
of cell suspension was mixed with 100 nM ASMase/NSMase siRNA into the
electroporation cuvette. HREC were then electroporated with the Nucleofactor program
M-030 (Amaxa Biosystems). The electroporated cells were maintained in supplemented
medium in 37°C/5% C02 incubator for 48 hours prior to TNFa (10 ng/ml) and IL-IB (5
ng/ml) treatment. To determine the efficiency of gene silencing, RNA was extracted from
100 nM siRNA treated HREC and used as template for real time-PCR as described (49).
ASMase/ NSMase siRNA treatment of HREC induced gene silencing of 94 % for

ASMase and 86 % for NSMase.

3.9. Quantitative Real Time-Polymerase Chain Reaction (qRT-PCR)

Total RNA was extracted from HREC with or without DHA pretreatment(50). Specific
primers for each gene were designed using IDT DNA PrimerQuest software (Coralville,
IA): for NSMase: forward —GAGCGCAATCATGGCAGT and reverse —CCCACTGA
ACCAGTCACCAT; for ASMase: forward —CAACCTCGCGCTGAAGAA and reverse-—
TCCACCATGTCATCCTCAAA. First strand cDNA was synthesized using the
SuperScript II RNase H-Reverse Transcriptase (Invitrogen Carlsbad, CA). Synthesized
cDNA was mixed with 2x SYBR Green PCR Master Mix (Invitrogen Carlsbad, CA) and

different sets of gene-specific forward and reverse primers, and then subjected to real-

74

time PCR quantification using the AB] PRISM 7900 Sequence Detection System
(Applied Biosystems). All reactions were performed in triplicates. The relative amounts
of mRNAs were calculated by using the comparative CT method (User Bulletin #2,
Applied Biosystems). Cyclophilin was used as a control and all results were normalized

to the abundance of cyclophilin mRNA.

3.10. Isolation of Caveolae/Lipid Rafts Membrane Domains

Isolation of caveolae/lipid raft-enriched detergent-resistant membrane domains were
prepared using a slightly modified sucrose gradient ultracentrifugation protocol(47).
Briefly, 5 X 106 HREC were washed twice with cold PBS, lysed in 0.8 ml MNE buffer
(25 mM MES, pH 6.5, 0.15 M NaCl, and 5 mM EDTA) containing 1% Triton X-100,
fresh protease and phosphatase inhibitors, and kept on ice for 20 minutes. Lysates were
then homogenized with 10 strokes of a tight fitting Dounce homogenizer and spun down
at 4000g at 4°C for 10 minutes. Supernatant (0.8 ml) was mixed with the same volume of
80% sucrose prepared in MNE buffer and was placed at the bottom of an ultracentrifuge
tube. Then 1.6 ml of 30% sucrose and 0.8 ml of 5% sucrose were overlaid on top of the
sample to form a 5 % to 30 % discontinuous sucrose gradient. After 18 hours of
centrifugation at 200,000 g and 4°C in a swinging bucket rotor, 0.4 ml samples were
collected carefully from the top of each fraction. A band confined to fractions 2 through 4

was designated as caveolae/lipid raft—enriched membrane domains.

75

3.11. Isolation of Plasma Membrane Domains

Cells were lysed in hypotonic buffer (10 mM HEPES, pH 7.4, 1 mM EDTA, 1 mM
MgC12 freshly added with protease inhibitors) for 30 minutes on ice, then homogenized
with 20 strokes in a nonstick (Teflon; DuPont, Wilmington, DE) glass homogenizer and
Spun down at 500g for 5 minutes. Supematants were subjected to centrifugation (16,900g,

60 minutes, 4°C) and pellets were collected as plasma membrane—enriched fractions.

3.12. Lipid Extraction

3 x 106 HREC were washed twice with cold PBS, collected in 500 uL cold 40%
methanol, and then homogenized in a glass homogenizer on ice. The cell homogenate
was transferred to a screw-cap glass tube and lipids were extracted as previously
described (51). The samples were dried under a stream of nitrogen and re-extracted twice
to remove non-lipid contaminants. The lipid film was dried under nitrogen, then further
dried overnight in a speedvac, and resuspended in an appropriate amount of isopropanol :

methanol : chloroform (4 : 2 : 1) based on the protein concentration of each sample.

3.13. Mass Spectrometry Analysis

Synthetic lipid standards ( Sphingolipid Mix 1, Avanti) were added to lipid extracts used
in mass spectrometry analysis at 20 nmol/mg protein and 5 nmol/mg protein,
respectively. Prior to analysis, lipids were appropriately diluted in

isopropanol/methanol/chlorofonn (4 : 2 : 1 v /v/ v) containing 20 mM ammonium

76

hydroxide, centrifuged, then loaded into Whatrnan Multichem 96-well plates (Fisher
Scientific, Pittsburgh, PA) and sealed with Teflon Ultra-Thin Sealing Tape (Analytical
Sales and Services, Pompton Plains, NJ). Lipids were introduced to a Thermo model TSQ
Quantum Ultra triple quadrupole mass spectrometer (San Jose, CA) as previously
described (51). Ceramide (Cer) molecular species were monitored as their [M+H]+ ions
by tandem mass spectrometry (MS/MS) precursor ion scanning for the formation of
characteristic product ions at m/z 264.4 (Cer) in positive ionization mode. All MS/MS
spectra were acquired automatically for 5-10 minutes at a rate of 500 m/z sec'l by
methods created using Xcalibur software (Thermo, San Jose, CA). Correction for 13.C
isotope effects and quantitation of lipid molecular species against appropriate internal
standards was performed using the Lipid Mass Spectrum Analysis (LIMSA) software
(52) peak model fit algorithm in conjunction with an expanded user-defined database of
hypothetical lipid compounds. Ceramide molecular species were quantitated
ratiometrically by comparison with the peak area of a synthetic lipid internal standard.
The LIMSA software peak model fit algorithm was used as previously described for peak
finding and isotope correction to arrive at an "absolute" value for each identified lipid
species. All identified ceramide species were summed in order to quantify the total
ceramide amount in the samples. MS/MS collision energies, CID gas pressure, and Q1
and Q3 settings were optimized using synthetic lipid standards corresponding to the

individual lipid classes of interest.

77

3.14. Statistical Analysis

Data are expressed as the mean i SD. Factorial ANOVA with the post hoc Tukey test
(GraphPad PrismS, GraphPad Software, San Diego, CA) was used for comparing the data

obtained from independent samples. Significance was established at P<0.05.

78

4. Results

4.1. Dose Response of TNFa. and IL-16 Induced CAMS Expression in HREC

To determine the dose of cytokines necessary to induce an inflammatory response
in HREC, dose response curves for TNFa (0-20ng/ml) and IL-IB (0.5-10ng/ml) were
established using CAMS (ICAM-1 and VCAM-1) expression as a measure. The induction
of ICAM-1 and VCAM-l by TNFa and IL-10 was assessed by immunoblot analysis
(Figure 2.1). Based on these data, we used 10 ng/ml TNFa and 5 ng/ml IL-lB in the

further studies.

4.2. TNFa. and IL-IB Induce ASMase and NSMase Activation in HREC

The effect of proinflammatory cytokines TNFa and IL-1 [3 known to be increased
in diabetic eyes on ASMase and NSMase activity in HREC was first determined. Cells
were stimulated with either 5ng/ml lL-IB or 10 ng/ml TNFa for lSsec — 2min. The cells
were then collected for ASMase and NSMase activity analysis using the Amplex Red
Sphingomyelinase assay kit. IL-lB treatment increased ASMase activity in HREC as
early as 15 seconds of stimulation and ASMase activity remained significantly elevated
during 30 seconds of stimulation (Figure 2.2A). Similar results were obtained for
NSMase activity, where IL-lB stimulation induced a maximum activation after 45
seconds of treatment (Figure 2.2 C). HREC treatment with pro-inflammatory cytokine
TNFa also Significantly induced ASMase and NSMase activation after 45 seconds of

stimulation (Figure 2.2, B and D).

79

4.3. Inhibition of Cytokine-Induced ASMase and NSMase Activation by DHA in

HREC

DHA, the most abundant w3-PUFA in the retina, has been shown to have a
pronounced anti-inflammatory effect, inhibiting cytokine-induced CAMS expression in
HREC and endothelial cell activation (12). To determine the effect of DHA on basal
ASMase and NSMase activity, HREC were treated with DHA or linoleic acid (lipid
control) for 24 hours. Pretreatment of HREC with 100 uM of BSA-bound DHA
significantly decreased both ASMase (Figures 2.3 A, B, E) and NSMase (Figures 2.3 C,
D, F) mRNA expression and activity level. Pretreatment of HREC with 100 uM of BSA-
bound linoleic acid had no effect on ASMase or NSMase basal activity when compared to
vehicle control (BSA). We next determined the effect of DHA on TNFa and IL-lB
induced ASMase and NSMase activity. As shown in Figure 2.3, DHA, but not linoleic
acid pretreatment of HREC, significantly down regulated TNFa and IL-lB-induced
ASMase (Figure 2.3 A and B) and NSMase (Figure 2.3 C and D) activity. To validate the
effect of DHA on cytokine-induced ASMase and NSMase activation, HREC were treated
with 15 pM desiprarnine (53-54) and 25 uM GW4869 (55) prior to addition of
stimulatory agents. Inhibition of cytokine-induced ASMase activity by desipramine is
shown in Figure 2.4 A and inhibition of cytokine-induced NSMase activity by GW4869

is shown in Figure 2.4 B.

80

4.4. Effect of ASMase and NSMase Inhibition and/or Gene Silencing on TNFa and

IL-IB Induced CAMS Expression

To establish the role of ASMase and NSMase in cytokine-induced inflammatory
signaling in HREC, ASMase and NSMase activity was inhibited using desipramine
(Sigma) and GW4869 (Sigma), respectively. HREC were treated with 15 uM
desipramine (53-54) for one hour and 25 uM GW4869 (55) for 30 minutes prior to TNFa
(10 ng/ml) stimulation. Six hours after cytokine stimulation the activation of
inflammatory pathways was analyzed using ICAM-1 and VCAM-l protein expression
levels as a measure. As shown in Figure 2.5, inhibition of ASMase Significantly reduced
the TNFa induced adhesion molecules expression, similarly to results achieved with
DHA treatment. Inhibition of ASMase also reduced IL-1I3 induced ICAM-l expression
(Figure 2.11 A, B and C). Interestingly, inhibition of NSMase did not have a significant
effect on TNFa and IL-1 [3 induced ICAM-l expression, and although the effect on TNFa
induced VCAM-1 expression was significant, it was less pronounced that the effect of

ASMase inhibition or DHA.

Gene silencing of ASMase and NSMase (100 nM siRNA for ASMase and
NSMase) significantly decreased TNFa and IL-lB-induced ICAM-1 and VCAM-1
expression in HREC (Figure 2.6) as compared to untreated or control siRNA (100 nM)
treated cells. Similarly to the inhibitors study, ASMase gene silencing had more
pronounced inhibitory effect compared to NSMase gene silencing. No additive effect on
cytokine induced CAM expression was observed in the samples where both ASMase and

NSMase gene silencing was attained.

81

4.5. Inhibition of ASMase and NSMase Activity by DHA

To determine the time course of DHA effect on ASMase and NSMase activity,
HREC were treated with BSA-bound DHA, or BSA alone. First, the dose of DHA to be
used in this experiment was selected based on dose—response experiment for ASMase
(Figure 2.7 A) and NSMase (Figure 2.7 C) activity. The time course for DHA effect on
ASMase and NSMase activity in HREC showed a gradual decrease in ASMase (Figure
2.7 B) and NSMase (Figure 2.7 D) activity with an early time point of 1 hour and 2
hours, respectively, and a maximum effect on both enzymes activity after 24 hours of
DHA treatment. To determine whether the effect of DHA on sphingomyelinases activity
is due to a direct interaction with the enzymes, we performed the sphingomyelinase assay
in a cell-free system where 0.5 mU of bacterial sphingomyelinase (homolog of NSMase)
was incubated with 5 to 80 uM fatty acid (DHA, linoleic acid). In this cell fi'ee system
fatty acids were used without a carrier molecule. Free fatty acid concentration of 5 mM
without a carrier in cell free system corresponds to 25 mM in cell culture experiments
where fatty acids are complexed to BSA at 1:5 molar ratio. No fatty acids were added to
the control samples. We found no direct effect of DHA or linoleic acid on SMase activity
(Figure 2.8), strongly suggesting that direct binding of DHA to sphingomyelinases does

not play a role in the observed inhibitory effects.

4.6. Inhibition of ASMase and NSMase Activity by MCD

DHA incorporation into the fatty acyl chains of phospholipids in caveolae/lipid

rafts causes cholesterol displacement from these specialized membrane microdomains

82

(47). To prove that lipid composition of caveolae/lipid rafts controls sphingomyelinase
activity, we determined the effect of cholesterol depletion on ASMase and NSMase in
HREC. Caveolae/lipid raft fractions were isolated by sucrose discontinuous gradient
ultracentrifugation based on their insolubility in Triton-X-100 at 4°C. Caveolin-1, a
caveolae marker, was used to identify the lipid raft fractions in density gradient fractions
2 to 4 (Figure 2.9 A). MCD and DHA pretreatment of HREC dramatically decreased
cholesterol content of the raft fractions (Figure 2.9 B) when compared to cholesterol
content of caveolae/lipid raft fractions isolated from BSA and linoleic acid pretreated

HREC.

Pretreatment of HREC with 8 mM MCD for 30 min decreased NSMase activity
by 42%, similarly to DHA effect on NSMase activity (Figure 2.9 C). Cholesterol
replenishment with 25 pM water-soluble cholesterol reversed the inhibitory effect on
NSMase activity (Figure 2.9 C). NSMase activity was only assessed in the whole cell
lysate since NSMase is only localized to the plasma membrane and activity in the whole
cell lysate represents the quantification of NSMase activity in the plasma membrane

compartment.

MCD treatment of HREC had no effect on whole cell lysate ASMase activity and
cholesterol replenishment did not restore ASMase activity in DHA treated HREC (Figure
2.9 D). AS a significant portion of cellular ASMase is in lysosomal compartment that
would not be affected by MCD and cholesterol treatments used, we next analyzed
ASMase activity in plasma membrane fiaction. ASMase activity was significantly

decreased in plasma membrane fraction isolated from MCD treated HREC and

83

cholesterol replenishment in DHA treated cells restored ASMase activity in this cellular

compartment (Figure 2.9 E).

4.7. No Effect of Ceramide de Nova Synthesis Inhibition on TNFa and IL-18

Induced CAMS Expression in HREC

Ceramides can be synthesized in the cells at the membrane level from
sphingomyelin hydrolysis by ASMase and/or NSMase or in the endoplasmic reticulum
by de nova synthesis. We next determined whether ceramide de novo synthesis pathway
is involved in mediating pro-inflammatory cytokine signaling in HREC. Fumonisin
Blwas used to inhibit ceramide synthase, the enzyme that catalyzes the final step of de
nova pathway of ceramide synthesis. Cells were treated for 16 hours with 50 uM
Fumonisin B1 prior to TNFa (10 ng/ml) and IL-1 I3 (5 ng/ml) stimulation. The expression
levels of ICAM-1 and VCAM-l were analyzed by western blot (Figure 2.10, A and D).
Fumonisin B1 pretreatment of HREC did not affect TNFa and IL-10 induced adhesion
molecules expression. In contrast, as Shown above, inhibition of ASMase significantly
decreased TNFa and IL-lB induced ICAM-1 and VCAM-1 expression in HREC. These
data suggest that ceramide production at the membrane level, rather than total cellular
ceramide level, plays an essential role in modulating cytokine induced inflammatory
signaling in HREC. Decreased ceramide levels in Fumonisin B1 treated HREC were
confirmed by tandem mass spectrometry analysis. Inhibition of ceramide synthase by
incubation of HREC with Fumonisin Bl resulted in a 26.55i9.45% decrease in HREC

total ceramide levels, as expected (Figure 2.10 l).

84

TNFa (ng/mL)

 

0 1 5 10 20

ICAM-1 ...... ...-nag

49““

VCAM-1 «up ’ u _

Tubulin - — - — -

lL-1B (ng/mL)

 

0 0.5 1 5 10
ICAM-1 ...-Eggs
VCAM-1 If .q" u q
Tubulin W

 

 

Figure 2.1. Dose response of cytokine-induced CAMS expression in HREC. HREC
maintained in serum fi'ee media overnight were stimulated with 0-20 ng/ml TNFa and 0-
10 ng/ml IL-lB for six hours. The induction of ICAM-1 and VCAM-l by TNFa (A) and
IL-1 B (B) was assessed by immunoblot analysis. Equal amounts of protein were added to
each lane as confirmed by tubulin level.

85

 

 

ASMase
.E
,9
9 "3 a
Q 4:
8 g 1 n’ I
a, L ....
a ‘z3 a?!
E V
3:) 0,7 --
lL-1B 0 15 30 45 60 120
Time (seconds)
B' ASMase

1.3 I

‘

0.9e
TNF“ o 15 30 45 60120

 

ASMase activity/pg protein
(Normalized units)

 

Time (seconds)

Figure 2.2. Cytokine induced ASMase and NSMase activation in HREC. HREC
cultured on 100m plates were stimulated with 5 ng/mL IL-l B (A,C) and 10 ng/mL TNFa
(B,D) for 0-120 seconds. Each time point represents an individual plate. The cells were
collected for ASMase (A,B) and NSMase (C,D) activity analysis using the Amplex Red
Sphingomyelinase assay kit. IL-lB significantly induced ASMase (A) activation after 15
seconds of stimulation, with a maximal activation after 30 seconds of stimulation. TNFa
induced ASMase (B) activation after 45 seconds of stimulation.

86

 

 

NSMase
.s
.9
3 a 3-5 *
g s
e 8 2.5
.2 E : '.
3 o 1.5 . ”II .3. A
to Z - _.-l
2 V o E"
(D
2 0.5 . , ._ w”.
O 15 30 45 60 120
lL-1|3
Time (seconds)
D.
NSMase
.5 2-5
2 *
o A -P
5 £3 2 - T.
c: C
3 . i --.
.2 .u 1.5 -
g g 1 . . . . . . $3.. *5... §
2 V "
(I)
TNFa 0 15 30 45 60120

 

Time (seconds)

Figure 2.2 (continued). Cytokine induced ASMase and NSMase activation in HREC.
IL-lB (C) or TNFa (D) induced maximal NSMase activation after 45 seconds of
stimulation. Results are presented as mean i SD of 3 independent experiments. *P<0.05
compared to basal expression level.

87

P
e:

 

 

 

 

 

 

 

   

 

 

.E ASMase ASMase
o
m C
E- 3 1.5 - 1
g 8 .5
.2 E
‘65 E 1 r 1
m E 3‘
a a * '
cu g 0.5 < 0.5 '
2
a)
‘5 _ _O._e_~-_ .. p- 0_‘
: Stimulation - IL-1B
i I BSA l
[j Linoleic acidl
' i
E] DHA i
C. D.
NSMase NSMase
.E
a 3 2.5 #
9 A it
0- .-‘_9 2.5 A f
c) C T 2 '
1 D
s v 2
._ cu 1.5 <
.2 .5
4g E 1.5
a 1
C?) 0.5 . 0.5 ‘
z .
I----z______,- . ,0,” - . - . -_ : * ,. _ , Q
‘1 Stimulation - lL-1B -

Figure 2.3. Inhibition of cytokine induced ASMase and NSMase activation, as well
as ASMase and NSMase expression by DHA in HREC. HREC treated with 100 pM
fatty acid (DHA or linoleic acid) complexed to BSA, or BSA alone (vehicle control)
overnight (24 hours) were stimulated with either 5 ng/mL lL-lB (A) for 30 seconds to
assess the activation of ASMase and for 45 seconds to assess the activation of NSMase;

88

F"
'51

 
 
  

 

 

 

 

ASMase NSMase

g 0.00003 0.0003 -

'5

2 0.00006 i

g. i 0.0002 A

0 l

3 0.00004 3

g l 0.0001 —

g 0.00002 i

r— 0 . __ 0 - __

 

I BSA
E] DHA

 

 

 

Figure 2.3 (continued). Inhibition of cytokine induced ASMase and NSMase
activation, as well as ASMase and NSMase expression by DHA in HREC. HREC
treated with 100 pM fatty acid (DHA or linoleic acid) complexed to BSA, or BSA alone
(vehicle control) overnight (24 hours) were stimulated with with 10 ng/mL TNFa for 45
seconds to assess both ASMase and NSMase activation. The cells were then collected for
ASMase (A,B) and NSMase (C,D) activity analysis using the Amplex Red
Sphingomyelinase assay kit. DHA decreased both ASMase (A,B) and NSMase (C,D)
basal activity and down-regulated cytokine-induced sphingomyelinases activity. RT-PCR
analysis of ASMase (E) and NSMase (F) expression in HREC with or without DHA
pretreatment. Results are presented as mean i SD of 3 independent experiments. *P<0.05
compared to basal expression level in BSA and linoleic acid treated HREC, #P<0.05
compared to non-stimulated HREC, IP<0.05 compared to cytokine stimulated HREC.

89

ASMase

 

 

 

 

 

ASMase activity/pg protein
(Normalized units)

BSA
BSA f;.:-_Ii<'ig,g.g§-*::_.:.;‘?ng;._-1'73,
DHA

D ir min £3)?“
\ esp a e

 

k
B. Y
IL-1B
3 _, NSMase

ASMase activity/pg protein
(Normalized units)

 

 

IL-10

Figure 2.4. Inhibition of ASMase and NSMase cytokine-induced activation by DHA
and sphingomyelinases inhibitors. HREC treated with 100 uM DHA complexed to
BSA overnight and inhibitors for ASMase (Desipramine, 15 BM for 1 hour) or NSMase
(GW4869, 25 BM for 30 min) down-regulated IL-lB-induced ASMase (A) and NSMase
(B) activity below the basal level, respectively. Results are presented as mean 1- SD of 3
independent experiments. #P<0.05 compared to basal expression level in BSA treated
HREC, 'P<0.05 compared to IL-lB stimulated, HREC.

90

 

BSA + + + + + +
TNFa - + + + + +
DHA - - + - - -
G W 4869 - - - + - +
Desipramine - - - - + +

 

 

Cyclophilin M

[CAM-1

lCAM-1/ thlaophilin
(Arbitrary units)

     

Desipramine _—4 *

 

<1: <0 0’
(<15 :1: g 5%
In 0 Eco
E g?
(D 9.;
:00
8+
TNFa

Figure 2.5. Decrease in TNFa induced CAMS expression by ASMase and NSMase
inhibition in HREC.

91

VCANL1

#3:

ICAM-1/ Cyclophilin
(Arbitrary units)

 

 

 

 

ores-crooo

 

*
ill * *
(D a) 0)
<75 <75 E 3 .5 58
m m D g E Egg
<0 E
<9 a .93
8 o f
TNFa

Figure 2.5 (continued). Decrease in TNFa. induced CAMs expression by ASMase
and NSMase inhibition in HREC. HREC were treated with 100 uM DHA complexed to
BSA, or BSA alone (vehicle control) for 24 hours, or the inhibitors for ASMase
(Desipramine, 15 pM for 1 hour) or NSMase (GW4869, 25 M for 30 min) prior to 10
ng/mL TNFa stimulation. The induction of ICAM-1 and VCAM-l was assessed 6 hours
later by irnmunoblot analysis (A). DHA, as well as Desipramine Significantly reduced the
TNFa induced CAM expression. GW4869 significantly inhibited TNFa induced
VCAM-1 expression. Quantitative analysis of the data presented in panel (A) for ICAM-
1 (B) and VCAM-l (C) expression. The results are meaniSD of 4 independent
experiments. #P<0.05 compared to control cells; *P<0.05 compared to TNFa stimulated
cells.

92

siRNA Control ASMase

 

I“ M ASMase
-- Tubulin

siRNA Control NSMase

 

F.” m ‘ NSMase
‘fl Tubulin

Figure 2.6. Inhibition in cytokine-induced CAMS expression by ASMase and
NSMase gene silencing in HREC. ASMase (A) and NSMase (B) protein level in control
siRNA, ASMase siRNA and NSMase siRNA treated HREC was assessed by immunoblot
analysiS.Gene silencing of ASMase and NSMase (100 nM siRNA for ASMase and
NSMase) significantly decreased TNFa (C) and IL-lB (F) induced ICAM-1 and VCAM-
1 expression in HREC. In contrast, control siRNA treatment (100 nM) had no effect.
Quantitative analysis of the data presented in C and F is shown in D,E and G,H
respectively. Equal amounts of protein were added to each lane, as confirmed by tubulin
levels. Results are presented as meaniSD of 5 independent experiments. # P<0.05
compared to control cells; *P<0.05 compared to TNFa and IL-1 B stimulated cells.

93

 

electroporation + + + + + +
TNFa - + + + + +
siRNA control - - + - .. -
siRNA ASMase - - - + - +
siRNA NSMase - - - - + +

 

ICAM-1 7‘ "‘2:

VCAM-1 ...... 3 - ....” .. ...,
Tubulin - * _. III-fl ~ and

D.
[CAM-1
1.5~
ET;
3*:
:53
': z~
‘T 9
2.1:
i 0 0i i i III
siRNA

Control

ASMase
NSMase
ASMase
+NSMase

 

TNFa

Figure 2.6 (continued). Inhibition in cytokine-induced CAMS expression by ASMase
and NSMase gene silencing in HREC.

94

VCA M-1

*

VCAM-1/ Tubulin
(Arbitrary units)

     

 

 

. I 4—0
SIRNA g ‘2“ 2
O a) (£3
< +
TNFa
F.
electroporation + + + + + +
151,3 - + + + + +
siRNA control - - + - - -
siRNA ASMase - - - + - +
- - - - + +

siRNA NSMase

 

ICAM'1 ‘ I ~ ‘ my old ......"
VCAM-1 { on. .0 w-v- but... «.14....
TUbulin W

Figure 2.6 (continued). Inhibition in cytokine-induced CAMs expression by ASMase
and NSMase gene silencing in HREC.

95

 

[CAM-1
1.5-
=2 :5 # #
.o c 1
.3 3 u.
?E * T *
E 1‘:
5510-5
0 6 (Did); 0%
. I I ,1: (D 0) mm
SIRNA g g g g:
a)
0 ‘2 ‘2 2’2;
lL-1B
H.
VCAM-1
1.5
E 13
3 4:
.135 1
: E
I E *
E 20.5a
o <
> V

     

NSMase
ASMase
+NSMase

 

lL-1

E

Figure 2.6 (continued). Inhibition in cytokine-induced CAMS expression by ASMase
and NSMase gene silencing in HREC.

96

.3’
cu

ASMase . ASMase

e111.

1 000

50° *"'§°---i--“§'"* * 51*

 

 

 

 

o§§§

 

 

ASMase activity/pg protein
(Normalized units)

 

 

Li... ...} *
°°"§
BSA 50 75 100 - 2’ 5’ 10’ 1h 2h 4h 6h12h 24h
DHA (nM) DHA treatment
C. D
NSMase NSMase
C
'6
‘9' A 600
2, ‘3 40° E .. ..f..
g: 3 400 "t. °§....l. * i
g 0N, 2m * i... '%'-f *
'- = ‘ .’ ' .. T
g g m 200 l *
m 0 "mi
as Z
2 V 0 - o . ,
(g BSA 50 75 100 - 2’ 5’ 10’ 1h 2h 4h 6h 12h 24h
DHA (nM) DHA treatment

Figure 2.7. Inhibition of ASMase and NSMase activity by DHA. HREC treated with
100 M DHA complexed to BSA, or BSA vehicle control overnight (24 hours) were
collected for ASMase (A) and NSMase (D) activity analysis using the Amplex Red
Sphingomyelinase assay kit. The optimal dose of DHA to inhibit ASMase (A) and
NSMase (C) activity in HREC was 100 1.1M. The time course for DHA effect on ASMase
and NSMase activity in HREC showed a gradual decrease in ASMase (B) and NSMase
(D) activity with an early time point of 1 hour and 2 hours, respectively and a maximum
effect on both enzymes activity afier 24 hours of DHA treatment. The results are meaniSD
of 3 independent experiments. 'P<0.05 compared to BSA vehicle control.

97

1-4
1.2

.+.

 

99
moo

SMase actIVIty
(Arbitrary units)

0.4
0.2

SuM IOpM ZOuM 40uM 80uM

Fatty acid concentration

 

 

 

 

 

 

 

 

 

 

 

 

 

; I Control

[:1 DHA

 

|
{ [:] Linoleic j
Figure 2.8. No direct effect of DHA on sphingomyelinase (SMase) activity in a cell-
free system. To determine whether DHA decreases SMase activity through direct
interaction with the enzyme, we measured SMase activity in a cell free system, where 0.5
mU of bacterial SMase (provided in the Amplex Red Sphingomyelinase assay kit) were
incubated with 5 to 80 uM fatty acids (DHA, linoleic acid). A concentration of 20 “M
free fatty acid corresponds to a concentration of 100 pM fatty acid complexed to BSA in
5:1 molar ratio. No fatty acids were used for the control wells. The assay was performed
as described in the Methods. No effect of DHA or linoleic acid on SMase activity was

observed when compared to control.

98

Fractions

 

 

:3...‘ ' “-
Linoleic acid .: ......

DHA .. ._ ~-~
MCD a... «1

BSA

 

 

 

 

Caveolin-1
B.
Cholesterol
200
150 m

(Arbitrary units)

 

 

1oo , *
50 I m
o 2 _

BSALinoleic DHA MCD

Cholesterol level/ pg of protein

Figure 2.9. Effect of cholesterol depletion on ASMase and NSMase activity in
HREC. Caveolae/lipid raft fractions were isolated by sucrose discontinuous gradient
ultracentrifugation based on their insolubility in Triton-X-IOO at 4°C and identified by
western blotting as fractions 2-4 based on the caveolin-1 distribution (A). MCD (white
bars) and DHA (dark gray bars) pretreatment of HREC significantly decreased
cholesterol content of the caveolae fractions when compared to BSA (carrier control) and
linoleic acid (lipid control) (B).

99

Figure 2.9 (continued). Effect of cholesterol depletion on ASMase and NSMase
activity in HREC. NSMase activity was inhibited by cholesterol depletion caused by the
treatment with 100 uM DHA overnight (gray bar) or 8 mM MCD for 30 min (white
bar)(C). The inhibitory effect of DHA on NSMase activity was reversed by cholesterol
replenishment using 25 14M water-soluble cholesterol for 30 minutes (hatched gray bar,
C). The inhibitory effect of DHA on ASMase activity was not mimicked by MCD and
cholesterol replenishment did not restore ASMase activity in DHA treated HREC in

whole cell lysates (D).

NSMase activity/pg protein

ASMase activity/pg protein

(Normalized units)

(Normalized units)

NSMase activity

Whole cell lysate

     

 

 

 

 

400 7
l
300 4' .
l :2. .
200 t
1oo
| $1.}?
0 “/29“ , ,
BSA DHA DHA MCD
+cholesterol
ASMase activity

Whole cell lysate

. Jr

300

i
200 1;

  

100

 

 

 

 

BSA DHA DHA MCD

+cholestero|

100

ASMase activity

Plasma membrane

     

P1

BSA DHA DHA MCD

ASMase activity/pg protein
(Normalized Ul‘lltS)

 

 

 

+cholesterol

Figure 2.9 (continued). Effect of cholesterol depletion on ASMase and NSMase
activity in HREC. However, cholesterol depletion inhibited ASMase activity and
cholesterol replenishment reversed the inhibitory effect of DHA on ASMase activity in
the plasma membrane fraction (E). Results are presented as meaniSD of 3 independent
experiments. *P<0.05 compared to BSA control. #P<0.05 compared to DHA.

101

 

 

IL-l B - + + +

Fumonisin Bl - - + -

Desipramine - - - +
ICAM-1 '“'“ _Q-—
VCAM-1 . ~ ~ ......

Tubulin M

 

 

B. C.
ICAM-1 VCAM-1
c A15 ~ # # 1,5
a .2 #
g S 1 . 1‘. #
E s
g g 0.5 I 0.5 _, *
. o 1 a a _ 01
I I E .93) I . E .E
.E E .E E
a g a g
8 a g a
E: 0 ..t 0
lL-1B IL-1t3

Figure 2.10. No effect of de novo ceramide production pathway inhibition on
cytokine induced CAMs expression in HREC. Fumonisin B1 was used to inhibit
ceramide synthase, the enzyme that catalyzes the final step in the de novo pathway of
ceramide synthesis. Cells were treated for 16 hours with 50 uM Fumonisin Bl prior to
TNFa (10 ng/ml) and IL-IB (5 ng/ml) stimulation. The expression levels of ICAM-1 and
VCAM-l were analyzed by western blot.

102

 

 

TNFa - + + +

Fumonisin Bl - - + _

Desipramine - - _ +
ICAM-1 1. g. ‘ I a -

VCAM-1 .. a- mi: - ..

*

Tubulin - - - -

[CAM-1 VCAM-1
1.5

0'5 0.5

o 1

Protem/ Tubulin
(Arbitrary units)

 

O
i
Desipramine -—-t .

 

 

I I f E | I a g
E .E E
.2 .2 9
c c Q
0 o .-
g g 3
LL “- 0
TN Fa TN Fa

Figure 2.10 (continued). No effect of de novo ceramide production pathway
inhibition on cytokine induced CAMs expression in HREC. Representative blots are
shown in A and D, and quantitative analysis of 4 independent experiments is shown in
B,C,E and F. Fumonisin Bl pre-treatment of HREC did not affect TNFa and IL-lB
induced adhesion molecule expression. In contrast, inhibition of ASMase significantly
decreased TNFa and IL-lB induced ICAM-1 and VCAM-1 expression.

103

Control HREC

C"3T(c11t3:1r24:0)

      
     
     
 
 

  

   

 

 

    
    
  

   

 

 

" Cer . . 650
°\o (d18.1116.0) Cer _ _
V - (d18-1l24.1) Cer - .
8IOO 0:1“: 1,120) 538 6 48‘»... |Satan/25.0)
53 «(432) Cer ‘ _Cer(d18:1/22:0) 6(64)
5 (d18.1118.0) 622
g L 566
< .__A x 11
H.
Fumonisin 81 treated HREC
cer(d*11;1I24:0)

A German/16:0) 650 Ce
5 ,Cefidtsmzn) 538 Ceqmmm.” “(a r(d18:1l25.0)
8100 (LS) 6481 '(l.S.)
g 432 C9r(d18:1t22:0) “ 664
1:7 “Rattan/18:0) 622
3 566
.0
< ___r___________,

470 570 670

mlz

Figure 2.10 (continued). No effect of de novo ceramide production pathway
inhibition on cytokine induced CAMs expression in HREC. Ceramide levels
determined by tandem mass spectrometry precursor ion mode scanning for the
characteristic ceramide product ion at m/z 264.4 in control and Fumonisin Bl treated
HREC are presented in G and H respectively.

104

1 400
1 200
1 000

 

 

Ceramide level
(pmol/mg protein)

I Control

 

 

El Fumonisin B1

 

Figure 2.10 (continued). No effect of de nova ceramide production pathway
inhibition on cytokine induced CAMs expression in HREC. Quantification of
ceramide levels presented in 1 demonstrates significant decrease in ceramide levels in
Fumonisin B1 treated compared to control HREC. Results are presented as meaniSD of

3 independent experiments. *P<0.05 compared to cytokine stimulated level. #P<0.05
compared to control.

105

5. Discussion

Omega 3 PUFAS have been long accepted to modulate inflammatory processes
and are now widely used in the clinic as an adjuvant immunosuppressant in the treatment
of various diseases with an inflammatory component (56). In a recent article, Connor et
al. showed that increasing omega 3 PUF A retinal levels by diet or genetic means may be
of benefit in preventing retinopathy and the DHA protective effect was mediated, in part,
by suppression of retinal TNFa gene expression and protein level in a mouse model of
oxygen-induced retinopathy (5). We have previously shown that DHA suppresses
cytokine induced adhesion molecule expression in endothelial cells through cholesterol
depletion and displacement of important signaling molecules from caveolae/lipid rafts
(47). We now demonstrate that DHA treated HREC exhibit decreased basal and TNFa
and IL-IB induced activity of ASMase and NSMase enzymes. The time course showed
maximum effect of DHA on sphingomyelinase activity after more than 12 hours of
treatment. This effect can be explained by our previously published observation that
DHA incorporates into caveolae/lipid rafts phospholipids at a rate of 10% in 1.5 hours
and 90% in 24 hours and dramatically alters the lipid environment of these specialized

microdomains, decreasing their cholesterol content by approximately 70% (47).

Both ASMase and NSMase have been described to localize in the caveolae/lipid
raft compartment of various cell types. Although ASMase was initially described as a
strictly lysosomal enzyme because of its optimum activity at pH 4.5-5.0, Liu and
Anderson have described the plasma membrane form of ASMase in caveolae and they

have shown that IL-IB stimulates its activity in this compartment (27). Veldman et al.

106

described a NSMase in caveolae isolated from human skin fibroblasts where the enzyme
appeared to interact with the scaffolding domain of caveolin protein (57). NSMase has
two putative transmembrane domains and five palmitoylation sites (29) and thus likely
has affinity for caveolae/lipid rafts microdomains. Cholesterol depletion as well as
displacement of signaling molecules essential for sphingomyelinases activation from
caveolae/lipid rafts by DHA may be the mechanism of DHA inhibitory effect on ASMase
and NSMase activity. Indeed, in this study, cholesterol depletion using MCD decreased
NSMase activity to the level similar to that observed with DHA. In contrast to NSMase,
ASMase has a major intracellular lysosomal component. The function and control of
ASMase activity in the plasma membrane and in lysosomal compartments are likely to be
different. Thus, it is not surprising that whole cell ASMase activity, which represents
lysosomal and plasma membrane ASMase combined, was not affected by short term
MCD and cholesterol treatments the same way as NSMase, which has only plasma
membrane localization. Plasma membrane ASMase activity was however, significantly
decreased by MCD, and cholesterol replenishment in DHA treated cells restored ASMase

activity in the plasma membrane.

Pro-inflammatory cytokines TNFa and IL-IB have been shown to activate
cellular ASMase in various cell types (27, 58-60) and our results are in agreement with
these studies. We have found that IL-IB treatment of HREC induced rapid activation of
ASMase afier only 15 seconds of stimulation, while TNFa increased both ASMase and
NSMase activity after 45 seconds of stimulation. This rapid and transient time course of
the effect is in agreement with the fact that induction of the inflammatory response is

always transient and the extent of the initial event is always small as it is later amplified

107

several fold with each signal transduction step. For instance, as demonstrated in
numerous studies of the NF KB pathway, IKBa is phosphorylated within a minute from
cytokine stimulation and is dephosphorylated back to the control state by 5 min; in GPCR
signaling ERK phosphorylation occurs within 30 seconds of receptor activation and it is
back to normal level by 2 minutes. Indeed, a recent study described the importance of
transient versus sustained ERK phosphorylation in inducing different signaling pathways
(61). Moreover, sphingomyelinase literature is replete with demonstration of the very fast
(within a minute) and transient nature of sphingomyelinase activation, and the fact that it
is so fast is often cited to promote the notion that sphingomyelinase is an important first
responder (58-59, 62). Although the induction of sphingomyelinases is transient, the
effect of the sphingomyelinase activation, e.g., conversion of sphingomyelin to ceramide
and further activation of the NFKB pathway and inflammatory genes, such as ICAM-1
and VCAM-l, is long lasting. Indeed, in this study inhibition and/or gene silencing of
ASMase and NSMase decreased pro-inflammatory cytokine TNFa and IL-lB induced

adhesion molecules expression in HREC.

The majority of studies on sphingomyelinase activation have used as a model
system the p55 TNFa receptor, showing that this receptor has a modular structure, with
particular domains for different sphingomyelinases. The intracellular C-terminal part of
the p55 receptor, which contains the death domain (DD), appears to be necessary for
ASMase activation; the exact molecular mechanism is not known, but Schwandner et al.
have reported ASMase activation (seconds to minutes) in response to TNFa through
overexpression of TRADD (TNF receptor associated DD) and/or F ADD (Pas-associating

protein with DD) (63). Kronke et al. have defined a region of this receptor that is adjacent

108

to the death domain and is required for NSMase activation (64). This motif was named
the NSMase activation domain. The same authors have also described a novel protein
(factor associated with NSMase activation, FAN) that binds the NSMase activation
domain on the TNFa receptor and is required for NSMase activation (65). The
mechanism by which lL-lB stimulates ASMase and NSMase activity is not yet

elucidated.

Inhibition and/or gene silencing of ASMase and NSMase decreased pro-
inflammatory cytokine TNFa and IL-18 induced adhesion molecules expression in
HREC. Both ASMase and NSMase are specific enzymes with phospholipase C activity,
which hydrolyze the sphingomyelin phosphodiester bond to produce ceramide and
phosphocholine. The ability of ceramide to self-aggregate and their fusagenic properties
may mediate membrane microdomain reorganization into large platforms required for
receptor clustering (66-68). As TNFa and IL-IB have receptors that appear to be
localized in the caveolae/lipid rafts fractions (58, 68) and protein oligomerization is an
almost universal requirement for transmembrane signal transmission, release of
ceramides alters the dynamics of these rafts and may drive and amplify inflammatory

signal transduction processes.

Moreover, there is evidence that stimulation of the sphingomyelin pathway by
TNFa with sphingomyelinases-induced membrane ceramide leads to activation of
nuclear factor KB (NF-KB) and marked increase in nuclear NF-KB binding in human
leukemia (HL-60) cells (59). Similarly, activation of sphingomyelinases is shown to be

an important signaling system for lL—lB in the murine T helper cell, EL-4 (58), and IL-18

109

action in these cells is mediated through NF-KB activation (69). NF-KB is a major
transcription factor controlling the expression of an array of inflammatory response genes

including adhesion molecules (70).

There are two different pathways of ceramide generation in cells. The first takes
place at the membrane level by hydrolysis of sphingomyelin by sphingomyelinases; the
second is localized to the endoplasmic reticulum by de novo synthesis involving the
enzyme ceramide synthase. The results of this study demonstrate that plasma membrane
ceramide production by sphingomyelinases rather than de novo synthesis is important for

inflammatory signaling in HREC.

Several recent studies demonstrated that oxidized products of DHA formed either
through lypoxygenase (5, 71-72), cyclooxygenase (72-73) or non-enzymatic pathways
(74) possess potent anti-inflammatory properties. Although anti-inflammatory mediators
formed by DHA oxidation are not likely to explain the results of this study, they might
represent an important component of anti-inflammatory action of DHA in other cell types
or under different experimental conditions. Endothelial cells are rich in caveolae and
ASMase is preferentially expressed in endothelia (75), thus our findings may represent an

identification of an endothelial-specific anti-inflammatory mechanism of DHA action.

In conclusion, this study provides a novel mechanism for the anti-inflammatory
effect of DHA in the primary tissue affected by retinopathy, human retina] endothelial
cells. We demonstrated that DHA-induced cholesterol depletion from membrane
microdomain leads to inhibition of ASMase and NSMase activity. Sphingomyelinase

downregulation by inhibitors and/or gene silencing recapitulated the DHA protective

110

effect against cytokine-induced inflammation in HREC. Importantly, preclusion of
cytokine-induced adhesion molecules in HREC was more pronounced in HREC when
ASMase rather than NSMase inhibition/gene silencing was attained, supporting the
general hypothesis that ASMase may play a more important role in active reorganization
of plasma membrane microdomains and receptor clustering (28). Furthermore, Veldman
et al. showed that after TNF stimulation, NSMase exited the caveolae and was followed
by a significant increase in ceramide content in this compartment, suggesting that
ASMase might mediate sphingomyelin hydrolysis (57). Grassme et al. demonstrated that
cells derived from Niemann-Pick disease patients (with ASMase deficiency) were
defective in receptor clustering and signal transduction in response to Fas, underlining the
importance of ASMase in oligomerization of cell surface proteins and transmembrane
signal transmission. Moreover, vascular endothelium is well recognized to express 20
times as much ASMase as any other cell type in the body (76) and this large excess of
ASMase in endothelium may be involved in tissue remodeling and wound repair (77),
processes that normally require waves of endothelial cell apoptosis and tissue

neovascularization (78).

Therefore, in the next chapter we characterized ASMase expression and cellular
localization in HREC when compared to other retinal tissues affected by diabetic
retinopathy, such as human retinal pigment epithelial cells and human Muller cells.
Moreover, we tested the DHA effect on caveolar versus intracellular ASMase expression
and we analyzed the sphingolipid composition of caveolae/lipid microdomains in control,

DHA treated and/or cytokine stimulated HREC.

111

6. References

10.

Lecomte, M., Paget, C., Ruggiero, D., Wiemsperger, N., and Lagarde, M. 1996.
Docosahexaenoic acid is a major n-3 polyunsaturated fatty acid in bovine retinal
microvessels. J Neuroch em 66:2160-2167.

SanGiovanni, J.P., Chew, E.Y., Clemons, T.E., Davis, M.D., Ferris, F.L., 3rd,
Gensler, G.R., Kurinij, N., Lindblad, A.S., Milton, R.C., Seddon, J.M., et a1.
2007. The relationship of dietary lipid intake and age-related macular
degeneration in a case-control study: AREDS Report No. 20. Arch Ophthalmol
125:671-679.

Decsi, T., Minda, H., Hermann, R., Kozari, A., Erhardt, E., Burns, 1., Molnar, S.,
and Soltesz, G. 2002. Polyunsaturated fatty acids in plasma and erythrocyte

membrane lipids of diabetic children. Prostaglandins Leukot Essent Fatty Acids
67:203-210.

Futterrnan, S., and Kupfer, C. 1968. The fatty acid composition of the retinal
vasculature of normal and diabetic human eyes. Invest Ophthalmol 7:105-108.

Connor, K.M., SanGiovanni, J.P., Lofqvist, C., Aderman, C.M., Chen, J.,
Higuchi, A., Hong, S., Pravda, E.A., Majchrzak, S., Carper, D., et a1. 2007.
Increased dietary intake of omega-3-polyunsaturated fatty acids reduces
pathological retinal angiogenesis. Nat Med 13 :868-873.

McLeod, D.S., Lefer, D.J., Merges, C., and Lutty, GA. 1995. Enhanced
expression of intracellular adhesion molecule-1 and P-selectin in the diabetic
human retina and choroid. Am J Pathol 147:642-653.

Joussen, A.M., Murata, T., Tsujikawa, A., Kirchhof, B., Bursell, S.E., and
Adamis, AP. 2001. Leukocyte-mediated endothelial cell injury and death in the
diabetic retina. Am J Pathol 158:147-152.

Adamis, AP. 2002. Is diabetic retinopathy an inflammatory disease? Br J
Ophthalmol 86:363-365.

Limb, G.A., Chignell, A.H., Green, W., LeRoy, F., and Dumonde, DC. 1996.
Distribution of TNF alpha and its reactive vascular adhesion molecules in
fibrovascular membranes of proliferative diabetic retinopathy. Br J Ophthalmol
80:168-173.

Joussen, A.M., Poulaki, V., Le, M.L., Koizumi, K., Esser, C., Janicki, H.,
Schraermeyer, U., Kociok, N., Fauser, S., Kirchhof, B., et al. 2004. A central role

for inflammation in the pathogenesis of diabetic retinopathy. Faseb J 18:1450-
1452.

112

ll.

12.

13.

14.

15.

l6.

17.

18.

19.

20.

21.

Adamis, A.P., Miller, J.W., Bemal, M.T., D'Amico, D.J., Folkman, J., Yeo, T.K.,
and Yeo, K.T. 1994. Increased vascular endothelial grth factor levels in the

vitreous of eyes with proliferative diabetic retinopathy. Am J Ophthalmol
118:445-450.

Chen, W., Esselman, W.J., Jump, DB, and Busik, J.V. 2005. Anti-inflammatory
effect of docosahexaenoic acid on cytokine-induced adhesion molecule

expression in human retinal vascular endothelial cells. Invest Ophthalmol Vis Sci
46:4342-4347.

Miyawaki-Shimizu, K., Predescu, D., Shimizu, J., Broman, M., Predescu, S., and
Malik, AB. 2006. siRNA-induced caveolin-1 knockdown in mice increases lung

vascular permeability via the junctional pathway. Am J Physiol Lung Cell Mol
Physiol 290:L405-413.

Li, S., Song, KS, and Lisanti, M.F. 1996. Expression and characterization of
recombinant caveolin. Purification by polyhistidine tagging and cholesterol-
dependent incorporation into defined lipid membranes. J Biol Chem 271:568-573.

Murata, M., Peranen, J., Schreiner, R., Wieland, F ., Kurzchalia, T.V., and Simons,
K. 1995. VIP21/caveolin is a cholesterol-binding protein. Proc Natl Acad Sci U S
A 92:10339-10343.

Anderson, R.G. 1993. Caveolae: where incoming and outgoing messengers meet.
Proc Natl Acad Sci U S A 90: 10909-10913.

Lisanti, M.P., Tang, Z., Scherer, P.B., Kubler, E., Koleske, A.J., and Sargiacomo,
M. 1995. Caveolae, transmembrane signalling and cellular transformation. Mol
Membr Biol 12:121-124.

Anderson, R.G., and Jacobson, K. 2002. A role for lipid shells in targeting
proteins to caveolae, rafts, and other lipid domains. Science 296: 1821-1825.

Razani, B., and Lisanti, M.F. 2001. Caveolins and caveolae: molecular and
functional relationships. Exp Cell Res 271 :36—44.

Smart, B.J., Graf, G.A., McNiven, M.A., Sessa, W.C., Engelman, J.A., Scherer,
P.B., Okamoto, T., and Lisanti, MP. 1999. Caveolins, liquid-ordered domains,
and signal transduction. Mol Cell Biol 19:7289-7304.

Sotgia, F., Razani, B., Bonuccelli, G., Schubert, W., Battista, M., Lee, H.,
Capozza, F., Schubert, A.L., Minetti, C., Buckley, J.T., et al. 2002. Intracellular
retention of glycosylphosphatidyl inositol-linked proteins in caveolin-deficient
cells. Mol Cell Biol 22:3905-3926.

113

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

Glenney, J.R., Jr. 1989. Tyrosine phosphorylation of a 22-kDa protein is
correlated with transformation by Rous sarcoma virus. J Biol Chem 264:20163-
20166.

Edidin, M. 2003. The state of lipid rafts: from model membranes to cells. Annu
Rev Biophys Biomol Struct 32:257-283.

Mathias, S., Pena, L.A., and Kolesnick, RN. 1998. Signal transduction of stress
via ceramide. Biochem J 335 ( Pt 3):465-480.

Bilderback, T.R., Gazula, V.R., Lisanti, M.P., and Dobrowsky, RT. 1999.
Caveolin interacts with Trk A and p75(NTR) and regulates neurotrophin signaling
pathways. J Biol Chem 274:257-263.

Cremesti, A., Paris, F., Grassme, H., Holler, N., Tschopp, J., Fuks, Z., Gulbins,
E., and Kolesnick, R. 2001. Ceramide enables fas to cap and kill. J Biol Chem
276:23954-23961.

Liu, P., and Anderson, R.G. 1995. Compartmentalized production of ceramide at
the cell surface. J Biol Chem 270:27179-27185.

Marchesini, N., and Hannun, Y.A. 2004. Acid and neutral sphingomyelinases:
roles and mechanisms of regulation. Biochem Cell Biol 82:27-44.

Tani, M., and Hannun, Y.A. 2007. Neutral sphingomyelinase 2 is palmitoylated
on multiple cysteine residues. Role of palmitoylation in subcellular localization. J
Biol Chem 282:10047-10056.

Tani, M., Ito, M., and Igarashi, Y. 2007. Ceramide/sphingosine/sphingosine 1-
phosphate metabolism on the cell surface and in the extracellular space. Cell
Signal 19:229-237.

Grassme, H., Schwarz, H., and Gulbins, E. 2001. Molecular mechanisms of
ceramide-mediated CD95 clustering. Biochem Biophys Res Commun 284:1016-
1030.

Schissel, S.L., Jiang, X., Tweedie-Hardman, J ., Jeong, T., Camejo, E.H., Najib, J .,
Rapp, J.H., Williams, K.J., and Tabas, I. 1998. Secretory sphingomyelinase, a
product of the acid sphingomyelinase gene, can hydrolyze atherogenic

lipoproteins at neutral pH. Implications for atherosclerotic lesion development. J
Biol Chem 273:2738-2746.

Smith, E.L., and Schuchman, EH. 2008. The unexpected role of acid

sphingomyelinase in cell death and the pathophysiology of common diseases.
Faseb J 22:3419-3431.

114

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

Kolesnick, R. 2002. The therapeutic potential of modulating the
ceramide/sphingomyelin pathway. J Clin Invest 110:3-8.

Pierini, L.M., and Maxfield, FR. 2001. Flotillas of lipid rafts fore and aft. Proc
Natl Acad Sci U S A 98:9471-9473.

Shaul, P.W., and Anderson, R.G. 1998. Role of plasmalemmal caveolae in signal
transduction. Am J Physiol 275:L843-851.

Simons, K., and Ikonen, E. 1997. Functional rafts in cell membranes. Nature
387:569-572.

Grassme, H., Riethmuller, J., and Gulbins, E. 2007. Biological aspects of
ceramide-enriched membrane domains. Prog Lipid Res 46: 161-1 70.

Holopainen, J .M., Subramanian, M., and Kinnunen, PK. 1998. Sphingomyelinase
induces lipid microdomain formation in a fluid
phosphatidylcholine/sphingomyelin membrane. Biochemistry 37:17562-17570.

Kolesnick, R.N., Goni, F.M., and Alonso, A. 2000. Compartmentalization of
ceramide signaling: physical foundations and biological effects. J Cell Physiol
184:285-300.

Nurminen, T.A., Holopainen, J .M., Zhao, H., and Kinnunen, PK. 2002.
Observation of topical catalysis by sphingomyelinase coupled to microspheres. J
Am Chem Soc 124:12129-12134.

Bertrand, D. 1997. [Mental health and cultural issues: the return of Khmers from
France to Cambodia for healing purposes]. Sante 7:330-334.

Monks, C.R., Freiberg, B.A., Kupfer, H., Sciaky, N., and Kupfer, A. 1998. Three-
dimensional segregation of supramolecular activation clusters in T cells. Nature
395:82-86.

Ko, Y.G., Lee, J.S., Kang, Y.S., Ahn, J.H., and Seo, J.S. 1999. TNF-alpha-
mediated apoptosis is initiated in caveolae-like domains. J Immunol 162:7217-
7223.

Martin, M.U., and Wesche, H. 2002. Summary and comparison of the signaling
mechanisms of the Toll/interleukin-l receptor family. Biochim Biophys Acta
1592:265-280.

Busik, J .V., Mohr, S., and Grant, MB. 2008. Hyperglycemia-induced reactive

oxygen species toxicity to endothelial cells is dependent on paracrine mediators.
Diabetes 57:1952-1965.

115

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

Chen, W., Jump, D.B., Esselman, W.J., and Busik, J.V. 2007. Inhibition of
cytokine signaling in human retina] endothelial cells through modification of
caveolae/lipid rafts by docosahexaenoic acid. Invest Ophthalmol Vis Sci 48: 1 8-26.

Jump, D.B., Clarke, S.D., MacDougald, 0., and Thelen, A. 1993. Polyunsaturated
fatty acids inhibit Sl4 gene transcription in rat liver and cultured hepatocytes.
Proc Natl Acad Sci U S A 90:8454-8458.

Botolin, D., and Jump, DB. 2003. Selective proteolytic processing of rat hepatic
sterol regulatory element binding protein-l (SREBP-l) and SREBP-Z during
postnatal development. J Biol Chem 278:6959-6962.

Agardh, E., Gustavsson, C., Hagert, P., Nilsson, M., and Agardh, CD. 2006.
Modifying a standard method allows simultaneous extraction of RNA and protein,
enabling detection of enzymes in the rat retina with low expressions and protein
levels. Metabolism 55:168-174.

Lydic, T.A., Busik, J .V., Esselman, W.J., and Reid, GE. 2009. Complementary
precursor ion and neutral loss scan mode tandem mass spectrometry for the
analysis of glycerophosphatidylethanolamine lipids fiom whole rat retina. Anal
Bioanal Chem.

Haimi, P., Uphoff, A., Hermansson, M., and Somerharju, P. 2006. Software tools
for analysis of mass spectrometric lipidome data. Anal Chem 7 818324-833 1.

Heinrich, M., Wickel, M., Schneider-Brachert, W., Sandberg, C., Gahr, J.,
Schwandner, R., Weber, T., Saftig, P., Peters, G, Brunner, J., et al. 1999.

Cathepsin D targeted by acid sphingomyelinase-derived ceramide. Embo J
18:5252-5263.

Kolzer, M., Werth, N., and Sandhoff, K. 2004. Interactions of acid
sphingomyelinase and lipid bilayers in the presence of the tricyclic antidepressant
desipramine. F EBS Lett 559296-98.

Marchesini, N., Luberto, C., and Hannun, Y.A. 2003. Biochemical properties of
mammalian neutral sphingomyelinase 2 and its role in sphingolipid metabolism. J
Biol Chem 278:13775-13783.

Calder, P.C. 2006. n-3 polyunsaturated fatty acids, inflammation, and
inflammatory diseases. Am J Clin Nutr 83:1505S-1519S.

Veldman, R.J., Maestre, N., Aduib, G.M., Medin, J .A., Salvayre, R., and Levade,
T. 2001. A neutral sphingomyelinase resides in sphingolipid-enriched
microdomains and is inhibited by the caveolin-scaffolding domain: potential
implications in tumour necrosis factor signalling. Biochem J 355:859-868.

116

58.

59.

60.

61.

62.

63.

64.

65.

66.

67.

68.

Mathias, 8., Younes, A., Kan, C.C., Orlow, 1., Joseph, C., and Kolesnick, RN.
1993. Activation of the sphingomyelin signaling pathway in intact EL4 cells and
in a cell-free system by IL-1 beta. Science 259:519-522.

Yang, Z., Costanzo, M., Golde, D.W., and Kolesnick, RN. 1993. Tumor necrosis
factor activation of the sphingomyelin pathway signals nuclear factor kappa B
translocation in intact HL-60 cells. J Biol Chem 268:20520-20523.

Marathe, S., Schissel, S.L., Yellin, M.J., Beatini, N., Mintzer, R., Williams, K.J.,
and Tabas, I. 1998. Human vascular endothelial cells are a rich and regulatable

source of secretory sphingomyelinase. Implications for early atherogenesis and
ceramide-mediated cell signaling. J Biol Chem 273:4081-4088.

York, R.D., Yao, H., Dillon, T., Ellig, C.L., Eckert, S.P., McCleskey, E.W., and
Stork, P.J. 1998. Rapl mediates sustained MAP kinase activation induced by
nerve growth factor. Nature 392:622-626.

Dressler, K.A., Mathias, S., and Kolesnick, RN. 1992. Tumor necrosis factor-
alpha activates the sphingomyelin signal transduction pathway in a cell-free
system. Science 255:1715-1718.

Schwandner, R., Wiegmann, K., Bernardo, K., Kreder, D., and Kronke, M. 1998.
TNF receptor death domain-associated proteins TRADD and FADD signal
activation of acid sphingomyelinase. J Biol Chem 273:5916-5922.

Wiegmann, K., Schutze, S., Machleidt, T., Witte, D., and Kronke, M. 1994.
Functional dichotomy of neutral and acidic sphingomyelinases in tumor necrosis
factor signaling. Cell 78: 1005-1015.

Adam-Klages, 8., Adam, D., Wiegmann, K., Struve, S., Kolanus, W., Schneider-
Mergener, J ., and Kronke, M. 1996. FAN, a novel WD-repeat protein, couples the
p55 TNF-receptor to neutral sphingomyelinase. Cell 86:937-947.

Belka, C., Marini, P., Budach, W., Schulze-Osthoff, K., Lang, F., Gulbins, E., and
Barnberg, M. 1998. Radiation-induced apoptosis in human lymphocytes and

lymphoma cells critically relies on the up-regulation of CD95/Fas/APO-l ligand.
Radiat Res 149:588-595.

Boniface, J .J ., Rabinowitz, J .D., Wulfing, C., Hampl, J ., Reich, Z., Altman, J .D.,
Kantor, R.M., Beeson, C., McConnell, H.M., and Davis, M.M. 1998. Initiation of
signal transduction through the T cell receptor requires the multivalent
engagement of peptide/MHC ligands [corrected]. Immunity 9:459-466.

Natoli, G., Costanzo, A., Guido, F., Moretti, F., and Levrero, M. 1998. Apoptotic,

non-apoptotic, and anti-apoptotic pathways of tumor necrosis factor signalling.
Biochem Pharmacol 56:915-920.

117

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

Stylianou, E., O'Neill, L.A., Rawlinson, L., Edbrooke, M.R., Woo, P., and
Saklatvala, J. 1992. Interleukin 1 induces NF-kappa B through its type I but not
its type II receptor in lymphocytes. J Biol Chem 267: 15836-1 5841.

Monaco, C., and Paleolog, E. 2004. Nuclear factor kappaB: a potential therapeutic
target in atherosclerosis and thrombosis. Cardiovasc Res 61 :671-682.

Marcheselli, V.L., Hong, S., Lukiw, W.J., Tian, X.H., Gronert, K., Musto, A.,
Hardy, M., Gimenez, J.M., Chiang, N., Serhan, C.N., et al. 2003. Novel
docosanoids inhibit brain ischemia-reperfusion-mediated leukocyte infiltration
and pro-inflammatory gene expression. J Biol Chem 278143 807-43817.

Schwab, J.M., Chiang, N., Arita, M., and Serhan, ON. 2007. Resolvin E1 and
protectin D1 activate inflammation-resolution programmes. Nature 447:869-874.

Serhan, C.N., Hong, S., Gronert, K., Colgan, S.P., Devchand, P.R., Mirick, G.,
and Moussignac, R.L. 2002. Resolvins: a family of bioactive products of omega-3
fatty acid transformation circuits initiated by aspirin treatment that counter
proinflammation signals. J Exp Med 196:1025-1037.

Musiek, E.S., Brooks, J.D., Joo, M., Brunoldi, E., Porta, A., Zanoni, G., Vidari,
G., Blackwell, T.S., Montine, T.J., Milne, G.L., et al. 2008. Electrophilic
Cyclopentenone Neuroprostanes Are Anti-inflammatory Mediators Formed fiom
the Peroxidation of the {omega}-3 Polyunsaturated Fatty Acid Docosahexaenoic
Acid. J Biol Chem 283:19927-19935.

Kolesnick, R., and Fuks, Z. 2003. Radiation and ceramide-induced apoptosis.
Oncogene 22:5897-5906.

Garcia-Barros, M., Paris, F ., Cordon-Cardo, C., Lyden, D., Rafii, S., Haimovitz-
Friedman, A., Fuks, Z., and Kolesnick, R. 2003. Tumor response to radiotherapy
regulated by endothelial cell apoptosis. Science 300:1 155-1159.

Jensen, J .M., Schutze, S., Forl, M., Kronke, M., and Proksch, E. 1999. Roles for
tumor necrosis factor receptor p55 and sphingomyelinase in repairing the

cutaneous permeability barrier. J Clin Invest 104:1761-1770.

Chavakis, E., and Dimmeler, S. 2002. Regulation of endothelial cell survival and
apoptosis during angiogenesis. Arterioscler T hromb Vasc Biol 22:887-893.

118

Chapter III

Docosahexaenoic Acid Regulates Caveolar Acid Sphingomyelinase
Expression and Promotes Sphingomyelin Enrichment of Membrane

Microdomains in Human Retinal Endothelial Cells

1. Abstract

Retinal endothelial cells represent the resident vasculature affected by diabetic
retinopathy. In addition to the retinal vasculature, diabetes may also result in damage to
nonvascular retinal tissue; yet microangiopathy represents the key pathologic feature of
diabetic retinopathy. As demonstrated in the previous chapter, acid sphingomyelinase
(ASMase) represents a key molecule in the cytokine mediated cellular inflammatory
cascade. To test the hypothesis that docosahexaenoic acid (DHA) has an anti-inflammatory
effect in diabetic retina by modulating ASMase expression specifically in the retinal
endothelial cells, we used three retinal cell types known to play a role in diabetes-induced
pathology: human retinal endothelial cells (HREC), adult human retinal pigment epithelial
cells (HRPE) and human Muller cells (HMC). We established that ASMase has the highest
activity and protein level in HREC, when compared to HRPE and HMC.
Immunohistochemistry assessment of ASMase distribution demonstrated that, although
ASMase was ubiquitously expressed in the perinuclear intracellular compartment of all
three cell types, HREC was the only cell type with pronounced plasma membrane ASMase
expression. Moreover, DHA decreased ASMase activity only in HREC, when compared to

HRPE and HMC. Importantly, DHA displaced ASMase from the caveolae membrane

119

microdomains in HREC and inhibited pro-inflammatory cytokine IL-IB induced lCAM-l
expression. ASMase gene silencing in HREC recapitulated DHA effects on cytokine-
induced endothelial cell activation. The effect of DHA on membrane microdomain
sphingolipid composition in HREC was analyzed by tandem mass spectrometry. In support
of ASMase inhibition and displacement of ASMase from caveolae membrane
microdomains by DHA, we found a significant decrease in ceramide/sphingomyelin ratio
in the caveolae fraction isolated from HREC treated with DHA. We have previously
demonstrated that DHA prevents TNFa-induced ASMase activation and now we show that
DHA prevents the TNFa-induced increase in the ceramide/sphingomyelin ratio in caveolar
membrane microdomains. Taken together, these in vitro data provide a mechanism for the
anti-inflammatory effect of DHA in endothelial cells, which we show to occur through

regulation of caveolar ASMase expression.

120

2. Introduction

Diabetic retinopathy is recognized as a microvascular complication of diabetes
and there is evidence that diabetes compromises the survival of vascular cells (1). Very
early stage diabetic retinopathy is regarded as a low-grade chronic inflammatory disease
with leukocyte adhesion to the retinal vasculature, a process mediated by adhesion
molecules expressed on the endothelial cell surface, especially intercellular adhesion
molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-l) (2-4).
Proinflammatory cytokines, including TNFa and IL-lB are increased in diabetic eyes (5-
7) and induce upregulation of adhesion molecule expression in vascular cells (8). Retinal

endothelial cells become activated and then degenerate as diabetic retinopathy progresses

(9, 10).

Vascular endothelial cells are not the only cell type affected by diabetes in the
retina. Muller cells, the main glial cells of the retina, are also dysfunctional in diabetes.
Muller cells, analogous to brain astrocytes, synthesize factors that participate in formation
of tight junction (11) and maintain the proper functioning of the inner blood-retinal
barrier under normal conditions (11, 12); they also synthesize and store various grth
factors with trophic and regulatory functions for a variety of cell types in the retina (13).
Mizutani et al. provided evidence for selective biosynthetic modifications of Muller cells

isolated from diabetic human eyes (1 3).

Another retinal cell type affected by diabetes are the retinal pigment epithelial
(RPE) cells that form the outer blood—retinal barrier. The RPE separates the neural retina

from the fenestrated choriocapillaries and have a major role in controlling the flow of

121

solutes, nutrients and fluid from the choroidal vasculature to the outer retina (14, 15).

Diabetes induces important morphological and biochemical changes in RPE cells (16-18).

HREC contain lipid rafts and particular plasma membrane microdomains known
as caveolae (19, 20) that have important roles in various physiological processes,
especially in regulating vascular permeability (21) and signal transduction (22, 23).
Caveolae are stabilized by the caveolins (24-27) and caveolin-1 is the primary structural

protein in endothelial cells caveolae (28).

Caveolae/lipid rafts microdomains are active assemblies of cholesterol,
sphingolipids (sphingomyelin, ceramide) and glycerophospholipids (29). Sphingolipids
are active components of membrane microdomains structure and are recognized to act as
messengers in signaling pathways mediating inflammation, apoptosis, cell differentiation
and proliferation (30). Sphingomyelinases are important early responders in inflammatory
cytokine signaling that catalyze the hydrolysis of sphingomyelin to the bioactive second
messenger ceramide (31-33). Acid sphingomyelinase (ASMase) is considered a major
candidate for mediating stress-induced production of ceramide. ASMase is best known as
a lysosomal enzyme where it plays an important role in lipid storage and sphingomyelin
metabolism. Recent studies, however, identified an important role of ASMase in
ceramide-mediated signal transduction. This unexpected function involves the
translocation of ASMase from intracellular compartments to the outer leaflet of cell
membranes. Hydrolysis of sphingomyelin into ceramide initiates a reorganization of the
membrane which facilitates formation and coalescence of lipid microdomains (34, 35).

Specifically, ASMase activity was localized to caveolae membrane microdomains (3 3).

122

Caveolae are enriched in sphingomyelin (26, 36-39) and thus represent a target
for sphingomyelinase translocation and an important pool for ceramide production.
Ceramide generation within the outer leaflet of the plasma membrane facilitates receptor
clustering and enhances inflammatory signal transduction by several receptors that
mediate the pathological changes typically associated with diabetic retinopathy, such as

IL-lB and TNFa (40, 41).

In this study, we examined a potential anti-inflammatory mechanism of DHA in
HREC through differential regulation of caveolar versus lysosomal ASMase localization
and promotion of sphingomyelin-enriched membrane microdomain formation with

important consequences on cellular inflammatory signaling.

123

3. Materials and Methods

3.1. Reagents and Antibodies

DMEM and F12 culture medium, antibiotics, fetal bovine serum and trypsin were
obtained from Invitrogen (Carlsbad, CA). Amplex Red Sphingomyelinase Assay Kit,
NuPAGE Novex 10% Bis- Tris gels, (Carlsbad, CA). GW4869 was purchased from
Calbiochem (San Diego, CA). Commonly used chemicals and reagents were purchased
fi'om Sigma (St. Louis, MO). Caveolin-1 and LAMP] antibodies (BD Bioscience, San
Jose, CA), ICAM-l antibody (Santa Cruz Biotechnology, Santa Cruz, CA) were used.
ASMase antibody was a generous gift from Dr. Richard Kolesnick. TNFa and IL-IB
were from R&Dsystems (Minneapolis, MN). ON-TARGET plus SMART pool SMPDl
(ASMase) and ON-TARGET plus siCONTROL (non-targeting pool, against luciferase)
were purchased from Dharmacon (Chicago, IL). For lipid extraction and mass
spectrometry, all solvents used were HPLC grade. Methanol (MeOH), and water were
purchased from J.T. Baker (Phillipsburg, NJ). Ammonium hydroxide (NH4OH) and
chloroform (CHC13) were from EMD Chemicals (Gibbstown, NJ). Isopropanol was from
Fisher Scientific (Pittsburgh, PA). Lipid standards were obtained from Avanti Polar

Lipids (Alabaster, AL).

3.2. Cell Culture

Primary cultures of HREC, HRPE and HMC were prepared from postmortem human
tissue (National Disease Research Interchange, Philadelphia, PA) cultured as previously

described (42). Passages 1 to 5 were used in the experiments. For experimental treatment,

124

cells were transferred to serum-free medium for 14 to 24 hours before stimulatory agents

were added. Only early passage cells were used in the experiments.

3.3. Fatty Acid Treatment

The method has been previously described in chapter 11. BSA was used as a carrier

control and linoleic acid, a (n6 polyunsaturated fatty acid was used as a lipid control.

3.4. Sphingomyelinase Assay

The method has been previously described in chapter 11.

3.5. Western Blot Analysis

The method has been previously described in chapter II.

3.6. Gene Silencing of ASMase

The method has been previously described in chapter II.

125

3.7. Isolation of Caveolae/Lipid Rafts Plasma Membrane Microdomains

a. The detergent-based method of caveolae/lipid raft isolation from HREC was
previously described in chapter 11. b. Non—detergent method of isolation of caveolae/lipid
rafts. Caveolae/lipid raft-enriched membrane microdomains were prepared using a
slightly modified non-detergent OptiPrep gradient ultracentrifugation protocol (43).
Briefly, 5 x 106 HREC were washed twice with cold PBS, collected into the base buffer
(20 mM Tris-HCl, pH 7.8, 250 mM sucrose) to which has been added 1 mM CaC12 and 1
mM MgC12, Cells were eluted by centrifugation for two minutes at 250 g and resuspended
in 0.33 ml of base buffer containing 1 mM CaC12 and 1 mM MgC12 and protease
inhibitors. The cells were then lysed by passage through a 22gx3” needle 20 times.
Lysates were centrifuged at 1,000g for 10 minutes. The resulting postnuclear supernatant
was collected and transferred to a separate tube. The pellet was again lysed by the
addition of 0.33 ml base buffer with divalent cations and protease inhibitors, followed by
sheering 20 times through a 22gx3” needle. After centrifugation at 1,000g for 10 minutes,
the second postnuclear supernatant was combined with the first. An equal volume (0.66
ml) of base buffer containing 50 % OptiPrep was added to the combined postnuclear
supematants and placed on the bottom of a 4 m1 centrifuge tube. A 2.4 ml gradient of 0 %
to 20 % OptiPrep in base buffer was poured on top of the lysate. Gradients were
centrifuged for 18 hours at 200,000g at 4 °C using a swinging bucket rotor and 0.4 ml
samples were collected carefully from the top of each fraction. A band confined to

fractions one and two was designed as caveolae/lipid raft- enriched membrane domains.

126

3.8. Isolation of Plasma Membrane Domains

The method has been previously described in chapter 11.

3.9. Lipid Extraction

Lipid extraction from whole HREC. 3 x 106 HREC were washed twice with cold PBS,
collected in 500 uL cold 40% methanol and homogenized in a glass homogenizer on ice.
The cell homogenate was transferred to a screw-cap glass tube and lipids were extracted
as previously described (44). The samples were dried under a stream of nitrogen and re-
extracted twice to remove non-lipid contaminants. The lipid film was dried under
nitrogen, then further dried overnight in a speedvac, and resuspended in an appropriate
amount of isopropanol : methanol : chloroform (4 : 2 : 1) based on the protein

concentration of each sample.

Lipid extraction from HREC caveolae/lipid raft fractions and plasma membranes. For
determination of sphingomyelin and ceramide content, plasma membranes and pooled
caveolin-l enriched OptiPrep fractions were subjected to lipid extraction with alkaline
hydrolysis of glycerophospholipids (45). Lipids were dried under a stream of nitrogen,
and re-extracted twice as outlined above for whole HREC lipid extracts to remove non-
lipid contaminants. Lipids were dried and resuspended as described above for whole

HREC lipids.

127

3.10. Mass Spectrometry Analysis

Synthetic lipid standards (Sphingolipid Mix 1, Avanti) were added to lipid extracts used
in mass spectrometry analysis at 20 nmol/mg protein and 5 nmol/mg protein,
respectively. Prior to analysis, lipids were appropriately diluted in isopropanol/methanol/
chloroform (4 : 2 : 1 v /v/ v) containing 20 mM ammonium hydroxide, centrifiiged, then
loaded into Whatrnan Multichem 96-well plates (Fisher Scientific, Pittsburgh, PA) and
sealed with Teflon Ultra-Thin Sealing Tape (Analytical Sales and Services, Pompton
Plains, NJ). Lipids were introduced to a Thermo model TSQ Quantum Ultra triple
quadrupole mass spectrometer (San Jose, CA) as previously described (44). Ceramide
(Cer) and sphingomyelin (SM) molecular species were monitored as their [M+H]+ ions
by tandem mass spectrometry (MS/MS) precursor ion scanning for the formation of
characteristic product ions at m/z 264.4 (Cer) and m/z 184 (SM) in positive ionization
mode. All MS/MS spectra were acquired automatically for 5-10 minutes at a rate of 500
m/z sec'l by methods created using Xcalibur software (Thermo, San Jose, CA).
Correction for '3 C isotope effects and quantitation of lipid molecular species against
appropriate internal standards was performed using the Lipid Mass Spectrum Analysis
(LIMSA) software (46) peak model fit algorithm in conjunction with an expanded user-
defined database of hypothetical lipid compounds. Ceramide and sphingomyelin
molecular species were quantitated ratiometrically by comparison with the peak area of
their synthetic lipid internal standard. The LIMSA software peak model fit algorithm was
used as previously described for peak finding and isotope correction to arrive at an
"absolute" value for each identified lipid species. All identified ceramide and

sphingomyelin molecular species were summed in order to quantify the total ceramide

128

and total sphingomyelin amount in the samples. The final result for each sample was
presented as ceramide/sphingomyelin ratio. MS/MS collision energies, CID gas pressure,
and Q1 and Q3 settings were optimized using synthetic lipid standards corresponding to

the individual lipid classes of interest.

3.11. Immunofluorescence Assay

Cells were cultured on cover slips. The next day, cells on cover slides were removed from
the media and rinsed in warm leBS three times. Then cells were fixed and
permeabilized in ice-cold methanol for 15 minutes at -20°C, washed with warm leBS
three times before blocking in a 2% BSA in leBS solution for one hour at 37°C, and
then incubated with primary anti-Caveolin—l, anti-ASMase and anti-LAMP] antibodies
and in a corresponding secondary antibody: Alexa Fluor 488 for Caveolin-land LAMPl
and Alexa Fluor 495 for ASMase. All the primary and secondary antibodies were used in
1:100 dilutions. The cells were washed three times (five minutes each) with warm leBS
before mounting the cover slips on the slides. ASMase and Caveolin-1 colocalization was
visualized with a confocal fluorescence microscope and measured with Olympus
Flroiew 1000 colocalization software. For images shown in Figure 3.2 B cells were
fixed in 2% paraformaldehyde in PBS and where indicated, cells were permeabilized by

ten minutes incubation in 0.1% TritonX-IOO.

129

3.12. Statistical Analysis

Data are expressed as the mean i SD. Factorial ANOVA with post hoc Tukey test
(GraphPad PrismS, GraphPad Software, San Diego, CA) was used for comparing the data
obtained from independent samples. T-test was used to compare data obtained from two

independent measurements. Significance was established at P<0.05.

130

4. Results

4.1. ASMase Expression and Activity in Human Retinal Cells

ASMase activity and expression level was assessed in three human retinal cell
cultures: human retinal endothelial cells (HREC), human adult retinal pigment epithelial
cells (HRPE) and human Muller cells (HMC). In these three retinal cell types, we found that
HREC have the highest ASMase activity (Figure 3.1 A) and protein levels (Figure 3.1 B,C)

when compared to HRPE and HMC.

4.2. ASMase Cellular Distribution in Human Retinal Cells

ASMase is ubiquitously expressed in the lysosomal compartment of all cell types.
Lysosomal ASMase plays an important role in sphingomyelin metabolism but is not
associated with inflammatory signaling. Recently endothelial cells were shown to secrete
copious amounts of ASMase in response to inflammatory cytokines and the
enzymatically active ASMase isofonn was found in the caveolae membrane microdomain
compartment of these endothelial cells (33). Caveolae are highly enriched in
sphingomyelin and 70% of total cellular sphingomyelin is found in these membrane
microdomains (26, 36-39). In order to hydrolyze plasma membrane sphingomyelin to
pro-inflammatory and pro-apoptotic ceramide, ASMase has to be present in the caveolae.
Therefore, we investigated the cellular distribution of ASMase in HREC, HRPE and
HMC. In HREC, we found that ASMase and caveolin-1, a marker for caveolae
microdomains, colocalize both at the membrane level and within the cytoplasm

56.38i5.85 % of ASMase staining colocalizes with caveolin-1 (Figure 3.2 A, upper

131

panel); however, we also observed a strong punctuate perinuclear staining negative for
caveolin-1, which likely represents lysosomal ASMase. To confirm that perinuclear
ASMase lies within the lysosomal compartment, lysosomal associated membrane protein
1 (LAMPl) was used as a marker for this cellular compartment. We found that
perinuclear ASMase closely associates with LAMPl (Figure 3.2 A, lower panel) as
indicated by the yellow color in the merged panel. Moreover, although ASMase was
ubiquitously expressed in the perinuclear intracellular compartment in HREC, HRPE and
HMC (Figure 3.2 B,D), HREC were the only cell type with pronounced plasma

membrane expression of ASMase (Figure 3.2 B,C)

4.3. DHA Effect on ASMase Activity in Human Retinal Cells

As we have previously shown, DHA-pretreatment decreases basal ASMase
activity in HREC. To compare the effect of DHA on ASMase activity on various retinal
cell types, we next determined ASMase activity in HRPE and HMC that were grown in
control media or media supplemented with DHA. DHA had no effect on ASMase activity

in HRPE (Figure 3.3 A) and HMC (Figure 3.3 B) when compared to control cells.

4.4. Downregulation of caveolar ASMase by DHA

Next we assessed ASMase colocalization with caveolin-1 in the caveolae/lipid
raft fi'actions isolated from HREC by sucrose discontinuous gradient ultracentrifugation.

Partial caveolae localization of ASMase was confirmed by western blot (Figure 3.4 A,C).

132

Moreover, DHA pretreatment of HREC prior to isolation of plasma membrane
microdomains shows downregulation of ASMase in caveolae microdomains (Figure 3.4

B,C) with no effect on the ASMase expression in the intracellular compartment.

4.5. Sphingomyelin Enrichment in Caveolae/Lipid Rafts by DHA

To address the effect of DHA on sphingolipid composition of caveolae/lipid rafts,
these microdomains were isolated by OptiPrep continuous gradient ultracentrifugation
then subjected to lipid extraction and lipid analysis by tandem mass spectrometry for
sphingomyelin and ceramide molecular species. Continuation of isolated lipid rafts was
achieved by showing enrichment of Caveolin-1 in density gradient fractions one and two.
We found a significant decrease in ceramide/sphingomyelin ratio in the caveolae/lipid
raft fi'actions isolated from HREC treated with DHA compared to BSA (vehicle control)
or linoleic acid (lipid control) (Figure 3.5A). Moreover, DHA pretreated HREC
prevented TNFa-induced increase in ceramide/sphingomyelin ratio (Figure 3.5 C). We
found no significant modification of ceramide/sphingomyelin ratio in the plasma
membrane domains isolated from control and DHA treated HREC (Figure 3.5 B). These
data show that DHA-induced ASMase downregulation in the caveolae compartment
results in formation of sphingomyelin-enriched caveolae/lipid rafts membrane

microdomains.

133

4.6. Gene Silencing of ASMase Prevents Proinflammatory Cytokine-Induced HREC

Activation

Treatment of HREC with SngmL IL-IB for 6 hours induced dramatic upregulation
of ICAM-1 protein expression (Figure 3.6 A,B). DHA significantly reduced cytokine-
induced ICAM-l protein expression in HREC (Figure 3.6 A,B). More importantly, ASMase
gene silencing recapitulated the effects of DHA on cytokine-induced adhesion molecules
expression in HREC, preventing the IL-lB-induced increase in ICAM-l protein level in

HREC (Figure 3.6 CD).

134

 

 

 

 

 

ASMase activtiy
1
.5 0.8
a)
E5
a 0.6 r
on
E
5 0.4 ~
E .
0.2 -‘
0 -1 --_ _. ___ ,
HREC HRPE HMC
B C.
.E ,
5 f
8 0.2 1
t— f
E ;
HREC HRPE HMC 8 Q1 * *
r _, v . \r .\ E, ;
ASMase.fl.“;§.'-:;..a; g, 5, < L
Tubulin mm 0 ,. . . _

HREC HRPE HMC

Figure 3.1. ASMase expression in human retinal cells. Three human retinal cell
cultures are shown: human retinal endothelial cells (HREC), human retinal pigment
epithelial (HRPE) and human Muller (HMC) cells. (A) ASMase activity in HREC,
HRPE and HMC. (B) ASMase expression in HREC, HRPE and HMC was determined by
immunoblot. (C) Quantitative analysis of the data in panel (B) for ASMase normalized
to tubulin level. The results are meaniSD of at least 3 independent experiments; *P<0.05
compared to HREC.

13S

Human retinal endothelial cells

 

 

Caveolin-1 l l ASMase l l Merged i
Nucleus Nucleus a
ar

/ ; Pefinucl
V
l

Membrane f i Membrane .

 

Perinuclear f

Perinuclear .
Membrane 7 Membrane

 

Figure 3.2. ASMase cellular distribution in human retinal cells. Three human retinal
cell cultures are shown: human retinal endothelial cells (HREC), human retinal pigment
epithelial (HRPE) and human Muller (HMC) cells. (A) ASMase, Caveolin-1 and LAMPl
colocalization in HREC was assessed by immunohistochemistry. There was 56.38i5.85
% colocalization of ASMase and caveolin-1 both at the plasma membrane level and in
the cytoplasm (upper panel). However, there was strong punctuate perinuclear staining,
negative for caveolin-1 that associates with LAMP] (lower panel), a marker for the
lysosomal compartment (yellow color in the Merged panel).

136

 

Non-permeabilized Permeabilized

‘____INucleus
O
UJ
0: /
I \

«Membrane NUCleUS
ii." \ Nucleus
CK
I
O
2
I

/
Nucleus

 

 

Figure 3.2 (continued). ASMase cellular distribution in human retinal cells. (B)
Immunofluorescent assessment of ASMase cellular distribution in non-penneabilized
(left) and permeabilized (right) HREC, HRPE and HMC.

137

ASMase staining in

non-permeabilized cells

 

 

 

 

'U

S

E E 20

8 cu g

(D

as g 10

‘0 0 .~.-.

9:3... 1:1 nd nd

5 s o - r

_S_.’ HREC HRPE HMC

Q

D.

ASMase staining in
permeabilized cells

'0

.8

E“ .15 2°

8 g S

as 5 10 ~

2.9 E- I g

.9 e

.E g 0 -. f

g HREC HRPE HMC

Figure 3.2 (continued). ASMase cellular distribution in human retinal cells. (C)
Quantification of ASMase staining in non-permeabilized cells corresponding to plasma
membrane localization of ASMase in HREC, HRPE and HMC (40 fields per
experiment). (D) Quantification of ASMase staining in permeabilized cells corresponding
to ASMase in the lysosomal compartment in HREC, HRPE and HMC (40 fields per
experiment). The results are meanISD of at least 3 independent experiments; nd= non-
deterrnined (red calibration bar = 27.5um, blue calibration bar = 55pm, white calibration
bar = llOum).

138

HRPE
300

 

100

ASMase activity/ ug protein
(Arbitrary units)

 

 

0 3111_

Control

 

HMC
150

 

100 “j

50,

ASMase activity/ pg protein
(Arbitrary units)

 

 

 

1--_,_-4_ 11.1.1.

can: DHA

Figure 3.3. The effect of DHA on ASMase activity in HRPE and HMC. ASMase
activity in (A) HRPE and (B) HMC treated for 18 hours with 100 uM DHA compared to
control. The results are meaniSD of three independent experiments.

139

A. Control

Ca veolin-1
ASMase

B- DHA treatment
1

Caveolin-1
ASMase

ASMase protein level
(Arbitrary units)

Figure 3.4. Downregulation of caveolar ASMase by DHA. (A-C) ASMase protein
expression level in caveolae fractions (fractions 3-4) versus non-caveolae fractions
(fractions 5-9) in control (A and C, white bar) and DHA (B and D, gray bar)-treated

HREC. The results are meaniSD of three independent experiments, +P<0.01 compared to

control.

 

Caveolae
(fractions 3-4) (fractions 5-9)

Fractions
2 3 4 5 6 7 8 9

- W ...-..

 

 

 

 

Fractions
2 3 4 5 6 7 8 9

"‘- “ 1......
“.1 .‘
‘. . _.

 

 

 

 

ASMase

 

Non-caveolae

140

Ceramide/sphingomyelin ratio

Membrane microdomains

 

 

 

 

 

 

0.08
.2
g 0.06 l
2*
g 0.04 ~
".5 I
E 0.02 I
Control DHA
B.
Ceramide/sphingomyelin ratio
Plasma membrane
,9 0.08 '
s 0.06 ~ I
2‘
g 0.04
L5
2 0.02 -‘

 

 

Control DHA

Figure 3.5. The effect of DHA on sphingomyelin and ceramide content of caveolae
membrane microdomain in HREC. (A) Decrease in ceramide/sphingomyelin ratio in
caveolae from DHA treated HREC compared to control. (B) No change in
ceramide/sphingomyelin ratio in total plasma membrane.

141

Ceramide/sphingomyelin ratio

Membrane microdomains

i

0.1 ~

 

0.05 %

 

Arbitrary units

   

cannot TNFa DHA+TNFa

G

Control HREC - membrane microdomains

12:0 Cer (l.S.) Ceramide

.3
0

24:1 Cer

24:0 Cer
16:0 C6,,H20 1630 Cer 12:0 Cerebroside (15.) 25:0 Cer US.)

l . l

480 500 520 540 560 580/ 600 620 640 660 680 700
mz

Relative abundance (%)
(II
o

1 6:0 SM Sphingomyelin

24:1 SM

2108M
26:1 SM

   
      
 
 

24:2 SM _

  

0 640 660 680 700 720 740 760 780 800 820 840
mlz

Relative abundance (%)

Figure 3.5 (continued). The effect of DHA on sphingomyelin and ceramide content
of caveolae membrane microdomain in HREC. (C) DHA prevents TNFa-induced
increase in ceramide/sphingo-myelin ratio in the caveolae. Tandem mass spectrum of
control HREC (D), DHA treated HREC (E), TNFa treated HREC (F) and DHA and
TNFa treated HREC (G) caveolae sphingomyelin species; and control HREC (H) and
DHA treated HREC (I) of plasma membrane sphingomyelin species by positive ion mode
PI m/z 184 after alkaline hydrolysis of glycerophospholipds.

142

I”

 

 

   
 
      
 

 

DHA treated HREC - membrane microdomains

 

 

 

 

g Ceramide
a 12:0 Cer (1.8.)
L5? 100 24:1 Cer
g 24:0 Cer
‘9 50 . 25:0 Cer (I.S.)
“ . 12:0 Cerebrosrde LS. .
g 16:0 Cer-H2016-O Cer ( ) :
*5 .1
6 l l . .l . .
0! 480 500 520 540 560 582/ 600 620 640 660 680 700
2
2\: 15:0 SM Sphingomyelin
8 100}
C -
g _
g 50; 24:1 SM
3 '120 SM (1 s ) 242 SM '240 SM
3 ' ' '1611 SM 18:0 SM ' 26:1 SM
59 .
g 640 660 680 700 720 74/0 760 780 800 820 840
m 2
F“ TNFa treated HREC - membrane microdomains
é Ceramide
0: 12:0 Cer (IS)
3 10' 24:1 Cer
53 24:0 Cer
C
{:3 5° 163° Cer 12:0 Cerebroside (I.S.) 253° 09' “-3)
.f’>_’ 16'0 cepHZO ' 223° Cer 26:0 Cer
E 1 . .l.. ..
carJ 480 500 520 540 560 582/2 600 620 640 660 680 700
$3
:1; 10 16:0 SM Sphingomyelin
C
8 24:1 SM
.5. 0
3‘; 12:0 sM (1.5.) ; ' 242 SM 24:0 SM
.3 l 16:1 SM < , 180 SM L 26:1 SM
% _ . - l 1 .4*‘ f." I‘M Al -1 1 gullmll .1 1.5L.
m 640 660 680 700 720 740 760 780 800 820 840

mlz

Figure 3.5 (continued). The effect of DHA on sphingomyelin and ceramide content
of caveolae membrane microdomain in HREC.

143

G. DHA and TNFa treated HREC - membrane microdomains

 

 
    
   
 

 

 

 

3 .

2, . Ceramide

8100 12.0Cer(l.8.) 2431C”

g 24:0 Cer

5 25:0 Cer (1.8.)

3 50 -

8 16-0 09' 12:0 Cerebroside (1.8.)

.QZ’ 16:0 Cer-H20 ‘ '

E

C?!) 480 500 520 540 560 580 600 620 640 660 680 700
m/z

015 S h'ngo el'

V . p I my In

§100 16.0 SM

(U

"U

C

.8

<0 50

a) 12:0 SM 1.8.

sf; | ( )16z18MHl l16:03M l l 24:1 SM

'0')

05 640 660 680 700 720 740 760 780 800 820 840

mlz

 

 
  
 
     
     
 

 

 

 

 

H' Control HREC - plasma membrane

24‘

V - Ceramide

g 100 12.0 Cer (1.8.) 24:1 Cer

"8“ 24:0 Cer

g 50 16 0 c 25:0 Cer (1.8.)

I er ' '

g 16:0 Cer-H20 12.0 Cerebrosrde (1.8.)

E

(“f 480 500 520 540 560 580/ 600 620 640 660 680 700
m Z

Ix;

V 16:0 8M ' '

8 1 003 Sphingomyelin

g -

g 50: 12'08M(18 ) 24:1SM

‘0 l ' ‘ ' 24:2 8M

3 14:0 SM 1618M l 17:0 SM 24:0 SM

.2 ‘: I - 22:0 8M

5 E . L. L 1 4h. .11., prSM . flu. in- . #1 11

6:9 640 660 680 700 720 740 760 780 800 820 840
mlz

Figure 3.5 (continued). The effect of DHA on sphingomyelin and ceramide content
of caveolae membrane microdomain in HREC.

144

[_DHA treated HREC - plasma membrane

 
    
   

  

 

 

 

 

 

§

V . Ceramide

g 100 12.0Cer(l.8.) 24:1Cer

'c ‘ 24ZOCer

C , _.

é 12:0Cerebroside(l.8.) 5 .’ 25:00er(|-S.)

0) 16:0 Cer 1‘

,.>_., 16:0 Cer-H20

s

0) I “I : a - ‘:

‘1 480 500 520 540 560 58,2, 600 620 640 660 680 700
Z

:5

E 100 16:0 8M Sphingomyelin

g 50 2418M

53 1208M (IS) 17:08M 24325”

> . . .

'8 L140s~1161 SM 18:0 SM 22:0 SM “.1340“ SM

6 a. - _ . 4.. . . -

‘1 640 660 680 700 720 74,0 760 780 800 820 840
"'12

Figure 3.5 (continued). The effect of DHA on sphingomyelin and ceramide content
of caveolae membrane microdomain in HREC. Tandem mass spectrum of control
HREC (D), DHA treated HREC (E), TNFa treated HREC (F) and DHA and TNFa
treated HREC (G) caveolae ceramide species; and control HREC (H) and DHA treated
HREC (I) of plasma membrane ceramide species by positive ion mode PI m/z 264. The
results are meaniSD of three independent experiments performed in triplicates, *P<0.05
compared to control, #P<0.05 compared to control and DHA treated and TNFa
stimulated HREC.

145

 

 

BSA + + +

DHA - - +

lL-1B - + +
lCAM-l d an...

Tubulin — - -1

B: ICAM-1

A
O
,1.

.1_._I__1_.___l__ .. ..

 

 

 

 

5 30
a is” /
8 ,9 //
.E 'o 20 , ‘
a E 10 /
g A
0 .____'I.'_ 244‘

v.
08
lL-1B

Figure 3.6. The effect of DHA and ASMase gene silencing on IL-lB-induced HREC
activation. (A) DHA or (B) ASMase siRNA treatment prevents IL-lB induced increase
in [CAM-1 expression in HREC. Random siRNA was used as control for ASM siRNA
experiments. The results are meaniSD of at least three independent experiments
performed in triplicates, *P<0.05 compared to control; #P<0.05 compared to IL-18
stimulated HREC, IP<0.05 compared to control.

 

146

 

 

Electroporation + + + +
lL-1B - + + +
siRNA control - - + -
siRNA ASMase - - - +
lCAM-‘l _ -— *-

Tubulin ---

 

 

 

 

 

 

D.
ICAM-1
60 g
T *
*
.E 1
2 ' , g.
E 3 40 fl}: “
9 .9 fit/3 ' .1
g ”g a;
--' .5 i;
E To 20 “ f/fi 2;
E /// =
8 = K
0 K: E
\ 0
I I $0 09
0

 

Figure 3.6 (continued). The effect of DHA and ASMase gene silencing on IL-lll-
induced HREC activation.

147

5. Discussion

The fluid mosaic model of the cell membrane structure proposed by Singer and
Nicholson in 1972 implied that membranes are in a disordered state with no major
selectivity (47). However, recently , an extensive literature has showed the existance of
microdomains that exist in a liquid-ordered phase within the membrane structure (48).
These membrane microdomains are highly enriched in sphingolipids (sphingomyelin,
ceramides) and cholesterol, are static within the membrane, but have the ability to
coalesce and rearrange in response to diverse stimuli (49). The generation of ceramide at
the plasma membrane level dramatically alters the properties of the cell membrane, since
ceramide molecules have the tendency to spontaneously self-associate (50, 51). It is
generally accepted that ceramide-rich membrane microdomains could primarily function
to reorganize receptors and various signaling molecules at the plasma membrane level in

order to facilitate and amplify cellular signaling processes via specific receptors (52).

Sphingomyelinases are sphingolipid hydrolyses that catalyze the breakdown of
sphingomyelin to ceramide and phosphorylcholine. Several isoforms of
sphingomyelinases have been identified and further distinguished by their catalytic pH
optimum, cellular localization, primary structure and co-factor dependence. Alkaline
sphingomyelinase activity is confined to the intestinal mucosa, bile and liver and does not
participate in signal transduction (36, 53, 54). Neutral sphingomyelinase (NSMase) and
ASMase are crucially involved in the pathophysiology of metabolic disorders (55) and

play an active role in cellular signaling (56).

ASMase is classically known to localize to the lysosomal compartment.

Lysosomal ASMase is ubiquitously expressed in all cell types. However, several studies

148

also identified a plasma membrane-associated form of ASMase in human fibroblasts (33)
and pulmonary vascular endothelial cells (57). Another study brought out evidence of the
existence of a small pool of zinc-independent ASMase in the plasma membrane
microdomains purified from NIH-3T3 L1 fibroblasts and PC12 cells and more
importantly, this pool of ASMase could undergo activation in response to neurotrophins,
whereas the lysosomal ASMase did not respond to activation by ligand (58). Various
immunohistochemistry studies have shown that cells exposed to different stress stimuli
exhibit a change in ASMase location from intracellular compartment to the cell surface
(59). Plasma membrane-associated ASMase, rather than intracellular ASMase, is shown
to be associated with inflammatory signaling (49), as ceramide generation within the
plasma membrane results in coalescence and reorganization of membrane microdomains
into large signaling platforms that facilitate receptor clustering and cellular signal

transduction (60, 61).

Endothelial cells are well recognized to be an important source of ASMase,
expressing as much as 20 times more ASMase as other mammalian cell types (62, 63).
Human coronary artery and human umbilical vein endothelial cells, as well as bovine and
murine aortic endothelial cells secrete copious amounts of ASMase in response to
inflammatory cytokines (63). Our study also shows that HREC exhibit high ASMase
activity and expression level when compared to other retinal tissues, including HRPE and
HMC cells (Figure 3.1 A,B,C). Moreover, in this study we demonstrate that it is this
plasma membrane ASMase that is a key element in the retinal inflammatory response in
HREC; HREC was the only cell type to demonstrate positive staining for the plasma

membrane form of ASMase, whereas HRPE and HMC presented only the classical,

149

intracellular form of ASMase. We have found in our study that 56.3 8i5.85 % of ASMase
colocalize with caveolin-1 both at the membrane and intracellular levels and 44 % may
represent the lysosomal form of ASMase. In endothelial cells, localization of ASMase to
the plasma membrane, and particularly to caveolae microdomains, increases substrate

availability for this enzyme.

Omega 3 PUFAS, and especially DHA, have been long accepted to modulate
inflammatory processes (64). We have previously shown that DHA pretreatment of
HREC decrease ASMase activity and gene expression (65). Our present study
demonstrates that DHA downregulates the ASMase protein level in the caveolae
microdomain compartment, with no effect on intracellular ASMase protein level.
Decreased ASMase expression in caveolae would profoundly impact the ability of HREC
to generate ceramide in this membrane microdomain. Although DHA has disparate
effects on the caveolar and lysosomal compartments, this differential sensitivity to DHA
is likely due to a higher exchange and/or degradation rate of the caveolar ASM, which is

dynamically regulated, compared to lysosomal ASM, which is constitutively expressed.

Indeed, regulation of caveolar ASMase expression has a profound impact on
membrane microdomains lipid composition, particularly on sphingomyelin and ceramide
content. As we have demonstrated in this study, downregulation of caveolar ASMase by
DHA induced a significant reduction in ceramide/sphingomyelin ratio and decreased the
ability of the proinflammatory cytokine TNFa to induce ceramide production in these
specialized membrane microdomains. However, as shown in the previous chapter, DHA
also downregulates the activity and cellular expression of NSMase, another critical

enzyme capable of sphingomyelin hydrolysis at the membrane level. Therefore, the DHA

150

effect on microdomain sphingolipid composition may be due to downregulation of both
ASMase and NSMase. Yet, inhibition and/or gene silencing of ASMase rather than
NSMase had a more pronounced effect on decreasing cytokine-induced adhesion
molecules expression in HREC. In addition, several studies have downplayed the role of
NSMase in ceramide-enriched platforms formation and transmembrane signal
transmission (66, 67). Taken together, our results suggest that ASMase could play a more
important role in cytokine-induced formation of ceramide-rich membrane microdomains

and amplification of inflammatory signaling in our cell culture system.

In conclusion, we demonstrated that HREC have the highest cellular ASMase
expression and protein level when compared to HRPE and HMC. Moreover, the plasma
membrane ASMase was expressed only in HREC. DHA downregulated caveolar ASMase
expression and promoted formation of sphingomyelin-enriched membrane microdomains.
These modified caveolae/lipid rafts have a decreased response to cytokines known to be
upregulated in diabetic retina (TNF-a and IL-IB) as confirmed by decreased cellular
adhesion molecules expression in HREC. Based on the results obtained in these in vitro
studies, we addressed in the following chapter the role of sphingomyelinases in retinal
vascular inflammation and vessel loss in diabetic retinopathy, using a streptozotocin
(STZ)-induced diabetes rat model; we also characterized the effect of DHA
supplementation on retinal sphingomyelinase expression, inflammatory status and

microangiopathy lesions in this type 1 diabetic rat model.

151

6. References

10.

Mizutani, M., Kern, TS, and Lorenzi, M. 1996. Accelerated death of retinal
microvascular cells in human and experimental diabetic retinopathy. J Clin Invest
97:2883-2890.

McLeod, D.S., Lefer, D.J., Merges, C., and Lutty, GA. 1995. Enhanced
expression of intracellular adhesion molecule-1 and P-selectin in the diabetic
human retina and choroid. Am J Pathol 147:642-653.

Joussen, A.M., Murata, T., Tsujikawa, A., Kirchhof, B., Bursell, S.E., and
Adamis, AP. 200]. Leukocyte-mediated endothelial cell injury and death in the
diabetic retina. Am J Pathol 158:147-152.

Adamis, AP. 2002. Is diabetic retinopathy an inflammatory disease? Br J
Ophthalmol 86:363-365.

Limb, G.A., Chignell, A.H., Green, W., LeRoy, F., and Dumonde, DC. 1996.
Distribution of TNF alpha and its reactive vascular adhesion molecules in

fibrovascular membranes of proliferative diabetic retinopathy. Br J Ophthalmol
80:168-173.

Joussen, A.M., Poulaki, V., Le, M.L., Koizumi, K., Esser, C., Janicki, H.,
Schraermeyer, U., Kociok, N., Fauser, S., Kirchhof, B., et al. 2004. A central role

for inflammation in the pathogenesis of diabetic retinopathy. Faseb J 18:1450—
1452.

Adamis, A.P., Miller, J .W., Bemal, M.T., D'Amico, D.J., Folkman, J ., Yeo, T.K.,
and Yeo, K.T. 1994. Increased vascular endothelial growth factor levels in the

vitreous of eyes with proliferative diabetic retinopathy. Am J Ophthalmol
118:445-450.

Chen, W., Esselman, W.J., Jump, DB, and Busik, J .V. 2005. Anti-inflammatory
effect of docosahexaenoic acid on cytokine-induced adhesion molecule

expression in human retinal vascular endothelial cells. Invest Ophthalmol Vis Sci
46:4342-4347.

Miyamoto, K., Khosrof, S., Bursell, S.E., Rohan, R., Murata, T., Clermont, A.C.,
Aiello, L.P., Ogura, Y., and Adamis, AP. 1999. Prevention of leukostasis and

vascular leakage in streptozotocin-induced diabetic retinopathy via intercellular
adhesion molecule-1 inhibition. Proc Natl Acad Sci U S A 96: 10836-10841.

Nozaki, M., Ogura, Y., Hirabayashi, Y., Saishin, Y., and Shimada, S. 2002.

Enhanced expression of adhesion molecules of the retinal vascular endothelium in
spontaneous diabetic rats. Ophthalmic Res 34: 1 58-164.

152

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Tout, S., Chan-Ling, T., Hollander, H., and Stone, 1. 1993. The role of Muller
cells in the formation of the blood-retinal barrier. Neuroscience 55:291-301.

Distler, C., and Dreher, Z. 1996. Glia cells of the monkey retina-—II. Muller cells.
Vision Res 36:2381-2394.

Mizutani, M., Gerhardinger, C., and Lorenzi, M. 1998. Muller cell changes in
human diabetic retinopathy. Diabetes 47:445-449.

Erickson, K.K., Sundstrom, J .M., and Antonetti, D.A. 2007. Vascular
permeability in ocular disease and the role of tight junctions. Angiogenesis
10:103-117.

Strauss, O. 2005. The retinal pigment epithelium in visual function. Physiol Rev
85:845-881.

Caldwell, R.B., and Slapnick, SM. 1989. Increased cytochrome oxidase activity
in the diabetic rat retinal pigment epithelium. Invest Ophthalmol Vis Sci 30:591-
599.

Grimes, P.A., and Laties, AM. 1980. Early morphological alteration of the
pigment epithelium in streptozotocin-induced diabetes: increased surface area of
the basal cell membrane. Exp Eye Res 30:631-639.

Tso, M.C., Cunha-Vaz, J .G., Shih, C.Y., and Jones, C.W. 1980. Clinicopathologic
study of blood-retinal barrier in experimental diabetes mellitus. Arch Ophthalmol
98:2032-2040.

Mora, R.C., Bonilha, V.L., Shin, B.C., Hu, J., Cohen-Gould, L., Bok, D., and
Rodriguez-Boulan, E. 2006. Bipolar assembly of caveolae in retinal pigment
epithelium. Am J Physiol Cell Physiol 290:C832-843.

Palade, GE. 1953. Fine structure of blood capillaries. J Appl Physiol 2421424—
1436.

Miyawaki-Shimizu, K., Predescu, D., Shimizu, J ., Broman, M., Predescu, S., and
Malik, AB. 2006. siRNA-induced caveolin-1 knockdown in mice increases lung
vascular permeability via the junctional pathway. Am J Physiol Lung Cell Mol
Physiol 2902L405-413.

Anderson, R.G. 1993. Caveolae: where incoming and outgoing messengers meet.
Proc Natl Acad Sci U S A 90: 10909-10913.

153

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

Lisanti, M.P., Tang, Z., Scherer, P.B., Kubler, E., Koleske, A.J., and Sargiacomo,
M. 1995. Caveolae, transmembrane signalling and cellular transformation. Mol
Membr Biol 122121-124.

Anderson, R.G., and Jacobson, K. 2002. A role for lipid shells in targeting
proteins to caveolae, rafis, and other lipid domains. Science 296:1821-1825.

Razani, B., and Lisanti, MP. 2001. Caveolins and caveolae: molecular and
functional relationships. Exp Cell Res 271:36-44.

Smart, B.J., Graf, G.A., McNiven, M.A., Sessa, W.C., Engelman, J.A., Scherer,
P.B., Okamoto, T., and Lisanti, MP. 1999. Caveolins, liquid-ordered domains,
and signal transduction. Mol Cell Biol 19:7289—7304.

Sotgia, F., Razani, B., Bonuccelli, G., Schubert, W., Battista, M., Lee, H.,
Capozza, F., Schubert, A.L., Minetti, C., Buckley, J .T., et al. 2002. Intracellular
retention of glycosylphosphatidyl inositol-linked proteins in caveolin-deficient
cells. Mol Cell Biol 22:3905-3926.

Glenney, J.R., Jr. 1989. Tyrosine phosphorylation of a 22-kDa protein is
correlated with transformation by Rous sarcoma virus. J Biol Chem 264:20163-
20166.

Edidin, M. 2003. The state of lipid rafts: from model membranes to cells. Annu
Rev Biophys Biomol Struct 32:257-283.

Mathias, S., Pena, L.A., and Kolesnick, RN. 1998. Signal transduction of stress
via ceramide. Biochem J 335 ( Pt 3):465-480.

Bilderback, T.R., Gazula, V.R., Lisanti, M.P., and Dobrowsky, RT. 1999.
Caveolin interacts with Trk A and p75(NTR) and regulates neurotrophin signaling
pathways. J Biol Chem 274:257-263.

Cremesti, A., Paris, F ., Grassme, H., Holler, N., Tschopp, J ., Fuks, Z., Gulbins,
E., and Kolesnick, R. 2001. Ceramide enables fas to cap and kill. J Biol Chem
276:23954-23 961 .

Liu, P., and Anderson, R.G. 1995. Compartmentalized production of ceramide at
the cell surface. J Biol Chem 270:27179-27185.

Gulbins, E., and Kolesnick, R. 2003. Raft ceramide in molecular medicine.
Oncogene 22:7070-7077.

Gulbins, E., and Li, PL. 2006. Physiological and pathophysiological aspects of
ceramide. Am J Physiol Regul Integr Comp Physiol 290:R1 1-26.

154

36.

37.

38.

39.

40.

41.

42.

43.

45.

46.

47.

48.

Kolesnick, R. 2002. The therapeutic potential of modulating the
ceramide/sphingomyelin pathway. J Clin Invest 110:3-8.

Pierini, L.M., and Maxfield, F .R. 2001. Flotillas of lipid rafts fore and aft. Proc
Natl Acad Sci U S A 98:9471-9473.

Shaul, P.W., and Anderson, R.G. 1998. Role of plasmalemmal caveolae in signal
transduction. Am J Physiol 27 5:L843-851 .

Simons, K., and Ikonen, E. 1997. Functional rafts in cell membranes. Nature
387:569-572.

Ko, Y.G., Lee, J.S., Kang, Y.S., Ahn, J.H., and Seo, J.S. 1999. TNF-alpha-
mediated apoptosis is initiated in caveolae-like domains. J Immunol 162:7217-
7223.

Martin, M.U., and Wesche, H. 2002. Summary and comparison of the signaling
mechanisms of the Toll/interleukin-l receptor family. Biochim Biophys Acta
1592:265-280.

Busik, J .V., Mohr, S., and Grant, MB. 2008. Hyperglycemia-induced reactive
oxygen species toxicity to endothelial cells is dependent on paracrine mediators.
Diabetes 57:1952-1965.

Macdonald, J .L., and Pike, L]. 2005. A simplified method for the preparation of
detergent-free lipid rafts. J Lipid Res 46:1061-1067.

Lydic, T.A., Busik, J.V., Esselman, W.J., and Reid, GE. 2009. Complementary
precursor ion and neutral loss scan mode tandem mass spectrometry for the
analysis of g1ycerophosphatidylethanolamine lipids from whole rat retina. Anal
Bioanal Chem.

Merrill, A.H., Jr., Sullards, M.C., Allegood, J.C., Kelly, S., and Wang, E. 2005.
Sphingolipidomics: hi gh-throughput, structure-specific, and quantitative analysis
of sphingolipids by liquid chromatography tandem mass spectrometry. Methods
36:207-224.

Haimi, P., Uphoff, A., Hermansson, M., and Somerharju, P. 2006. Software tools
for analysis of mass spectrometric lipidome data. Anal Chem 78:8324-8331.

Singer, S.J., and Nicolson, G.L. 1972. The fluid mosaic model of the structure of
cell membranes. Science 1 75 :7 20-73 1 .

Jacobson, K., Sheets, ED, and Simson, R. 1995. Revisiting the fluid mosaic
model of membranes. Science 268:1441-1442.

155

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

Schuchman, EH. 2009. Acid sphingomyelinase, cell membranes and human
disease: Lessons from Niemann-Pick disease. FEBS Lett.

Holopainen, J .M., Subramanian, M., and Kinnunen, PK. 1998. Sphingomyelinase
induces lipid microdomain formation in a fluid
phosphatidylcholine/sphingomyelin membrane. Biochemistry 3 7:1 7562-17570.

Kolesnick, R.N., Goni, F.M., and Alonso, A. 2000. Compartrnentalization of
ceramide signaling: physical foundations and biological effects. J Cell Physiol
184:285-300.

Grassme, H., Riethmuller, J., and Gulbins, E. 2007. Biological aspects of
ceramide-enriched membrane domains. Prog Lipid Res 46:161-170.

Duan, RD. 2006. Alkaline sphingomyelinase: an old enzyme with novel
implications. Biochim Biophys Acta 1761:281-291.

Pavoine, C., and Pecker, F. 2009. Sphingomyelinases: their regulation and roles in
cardiovascular pathophysiology. Cardiovasc Res 82:175-183.

Holland, W.L., and Summers, SA. 2008. Sphingolipids, insulin resistance, and
metabolic disease: new insights from in vivo manipulation of sphingolipid
metabolism. Endocr Rev 29:381-402.

Marchesini, N., and Hannun, Y.A. 2004. Acid and neutral sphingomyelinases:
roles and mechanisms of regulation. Biochem Cell Biol 82:27-44.

Yang, Y., Yin, J ., Baumgartner, W., Samapati, R., Solymosi, E.A., Reppien, E.,
Kuebler, W.M., and Uhlig, S. 2009. Platelet-activating factor reduces endothelial
NO production - role of acid sphingomyelinase. Eur Respir J.

Dobrowsky, RT. 2000. Sphingolipid signalling domains floating on rafts or
buried in caves? Cell Signal 12281-90.

Gulbins, E. 2003. Regulation of death receptor signaling and apoptosis by
ceramide. Pharmacol Res 47:393-399.

Grassme, H., Jendrossek, V., Riehle, A., von Kurthy, G., Berger, J ., Schwarz, H.,
Weller, M., Kolesnick, R., and Gulbins, E. 2003. Host defense against
Pseudomonas aeruginosa requires ceramide-rich membrane rafts. Nat Med 9:322-
330.

Gulbins, E., and Grassme, H. 2002. Ceramide and cell death receptor clustering.
Biochim Biophys Acta 1585:139-145.

156

62.

63.

64.

65.

66.

67.

Andrieu-Abadie, N., and Levade, T. 2002. Sphingomyelin hydrolysis during
apoptosis. Biochim Biophys Acta 1585: 1 26-134.

Marathe, S., Schissel, S.L., Yellin, M.J., Beatini, N., Mintzer, R., Williams, K.J.,
and Tabas, I. 1998. Human vascular endothelial cells are a rich and regulatable
source of secretory sphingomyelinase. Implications for early atherogenesis and
ceramide-mediated cell signaling. J Biol Chem 273:4081-4088.

Calder, P.C. 2006. n-3 polyunsaturated fatty acids, inflammation, and
inflammatory diseases. Am J Clin Nutr 83:1505S-1519S.

Opreanu, M., Lydic, T.A., Reid, G.E., McSorley, K.M., Esselman, W.J., and
Busik, J. 2010. Docosahexaenoic Acid Inhibits Cytokine Signaling in Human
Retinal Endothelial Cells by Downregulating Sphingomyelinases. Invest
Ophthalmol Vis Sci.

Grassme, H., Jekle, A., Riehle, A., Schwarz, H., Berger, J., Sandhoff, K.,
Kolesnick, R., and Gulbins, E. 2001. CD95 signaling via ceramide-rich membrane
rafts. J Biol Chem 276:20589-20596.

Veldman, R.J., Maestre, N., Aduib, G.M., Medin, J .A., Salvayre, R., and Levade,
T. 2001. A neutral sphingomyelinase resides in sphingolipid-enriched
microdomains and is inhibited by the caveolin-scaffolding domain: potential
implications in tumour necrosis factor signalling. Biochem J 355:859-868.

157

Chapter IV

Docosahexaenoic Acid Supplementation Prevents Upregulation of
Proinflammatory/Angiogenic Molecules and Vessel Loss in Diabetic Retina

Through Downregulation of Acid Sphingomyelinase

I . Abstract

Sphingomyelinases (SMases) are key regulatory enzymes of sphingolipid
metabolism, catalyzing hydrolysis of sphingomyelin to the proinflammatory and pro-
apoptotic messenger ceramide. In this study, we addressed the role of SMases in retinal
vascular inflammation and vessel loss in diabetic retinopathy, using a streptozotocin
(STZ)-induced diabetes rat model. Acid (ASMase), but not neutral (N SMase) SMase was
upregulated in the retinas with microangiopathy lesions. Proinflarnmatory/angiogenic
molecules were significantly increased in retina of STZ rats at one month and nine month
post STZ treatment. However, a docosahexaenoic acid (DHA)-enriched diet restored
ASMase in diabetic retina to the control levels and prevented upregulation of
inflammatory/angiogenic mediators’ gene expression, strongly demonstrating a DHA
protective effect against the development of an early inflammatory status in the diabetic
retina. More importantly, DHA prevented retinal vessel loss in the retina of STZ treated,
diabetic rats. In order to confirm that the protective effect of DHA in the retina is
mediated through vascular ASMase, we next used an ischemia/reperfusion (I/R) injury
model of retinopathy in rats. Immunohistochemistry analysis of normal rat retina

demonstrated that retinal vasculature preferentially express ASMase when compared the

158

other retinal tissues. Furthermore, rat retinal I/R specifically increased ASMase
expression in the vasculature, with little or no effect on ASMase expression in other
retinal cells. Notably, DHA supplementation prior to retinal injury prevented ASMase
upregulation in the retinal vasculature. Taken together, these in vivo data demonstrated
ASMase downregulation as a mechanism by which DHA exerts its protective effect

against retinal microvascular damage in diabetic retinopathy.

159

2. Introduction

Visual impairment among people with diabetic retinopathy is a major disability
with a profound impact on quality of life (1). Diabetic retinopathy (DR) represents the
ocular expression of diabetic microangiopathy. The natural history of DR has been
classified into an early, non-proliferative stage (background retinopathy) and a later,
proliferative stage. The early stage of DR has distinct vascular lesions that are
histologically characterized by the existence of saccular capillary microaneurysms, retinal
hemorrhage, venous dilation and beading, retinal lipid exudates, pericyte degeneration,
occlusion and degeneration of retinal capillaries (2-3). This early stage diabetic
retinopathy represents a persistent low-grade chronic inflammatory disease (4) that
involves leukocyte recruitment and adhesion to the retinal vasculature and up-regulation
of inflammatory genes. Adhesion molecules, especially intercellular adhesion molecule-1
(ICAM-l), are involved in leukocyte attachment and transmigration into the vascular
intima (5-6) that will result in early blood retinal barrier breakdown, capillary non-
perfusion, endothelial cell injury and death. Proinflammatory cytokines (TNFa, IL-IB,
IL-6 and VEGF) are increased in diabetic eyes and several inflammatory pathways are
activated in the very early stages of diabetic retinopathy (7-11).

Proinflammatory cytokines that are upregulated in diabetic eyes, including TNFa

and IL-lB have been shown to activate cellular sphingomyelinases in various cell types
(12-15), including retinal vascular endothelial cells, as presented in the previous chapters.
Sphingomyelinases are enzymes that catalyze the hydrolysis of sphingomyelin (ceramide
phosphocholine) to the bioactive lipid ceramide. Caveolae microdomains present on

endothelial cells are enriched in sphingomyelin and thus represent an important pool for

160

ceramide production. Ceramide generation within the outer leaflet of the plasma
membrane facilitates receptor clustering and enhances inflammatory signal transduction
by several receptors that mediate the pathological changes typically associated with
diabetic retinopathy, such as IL-IB and TNFa (16-17). As previously underlined, vascular
endothelial cells represent the target tissue affected by diabetic retinopathy and become
dysfunctional and degenerate due to diabetic metabolic dysregulation and inflammatory
stress. Moreover, vascular endothelial cells are considered to be a very rich source of
secreted ASMase and pro-inflammatory cytokines, especially IL-l, increase ASMase

secretion in in vivo studies (18).

The mechanisms that induce the proinflammatory state in the retina are not well
defined, but various studies have shown that hyperglycemia and dyslipidemia play an
important role. Hyperglycemia, a major causative factor in the development of diabetic
retinopathy, induces cytokine production by several distinct retinal cell types including
retinal pigment epithelium, Muller cells, astrocytes, microglia and pericytes, but
interestingly not by retinal endothelial cells. We previously demonstrated in vitro that
retinal endothelial cells increase the production of reactive oxygen species in response to
diabetes induced cytokine release by other retinal cells rather than directly to high
glucose, suggesting that in vivo diabetes-related endothelial injury in the retina is also due
to glucose-induced cytokine release and not a direct effect of high glucose on endothelial
cells (19). Retinal endothelial cells express TNF-a and IL-IB receptors and stimulation
with cytokines activate classic NF KB inflammatory pathways leading to increase

expression of inflammatory genes, including adhesion molecules ICAM-1 and VCAM-l.

161

In addition to hyperglycemia, Diabetes Control and Complications and
Epidemiology of Diabetes Interventions and Complications Trials (DCCT/EDIC)
demonstrated the strong association between dyslipidemia and development of diabetic
retinopathy. We have previously demonstrated that the retina has a unique fatty acid
composition exhibiting the highest level of (03-polyunsaturated fatty acids (PUFA),
especially DHA, in the body and that retinal fatty acid metabolism is distinct from the

metabolic changes that occur in diabetic plasma or diabetic liver (20).

Omega 3 PUFAS have the ability to modulate various biological processes
involved in retinal vascular inflammation, retinal capillary structure and integrity, and
retinal angiogenesis (21). We have previously shown that DHA suppresses cytokine
induced inflammatory signaling and activation of HREC (22). This pronounced anti-
inflammatory effect of DHA was mediated through a reduction in ASMase and NSMase
activity and expression (23). Importantly, total m3-PUFAs, and particularly DHA, are
reduced in the retina of human (24) and rat (20) diabetic eyes. Moreover, increasing the
retinal levels of (o3-PUFAs by genetic or dietary means has a beneficial effect in reducing

pathological retinal angiogenesis (21).

Activation of sphingomyelinases is an important early step in inflammatory
cytokine signaling (13-14) particularly in endothelial cells (15, 23). In this study, we
tested the hypothesis that downregulation of SMases by DHA may be protective against
endothelial activation, injury and death, and thereby preventing retinal vessel loss in a rat

model of type 1 diabetes.

162

3. Materials and Methods

3.1. Reagents and Antibodies

Crude trypsin was obtained from Difco (Detroit, MI)4.. Lyophilized pancreatic elastase
was purchased from Calbiochem (San Diego, CA). Streptozotocin (Sigma, St. Louis,
MO), Neutral Protamine Hagedom (NPH) insulin, experimental diets Dyets Inc.
(Bethlehem, PA), RT-PCR primers from IDT DNA Technologies (Coralville, IA) (Table
4.4 and Table 4.5). ASMase antibody was a generous gift from Dr. Richard Kolesnick.
VEGF antibody was purchased from Novus Biologicals (Littleton, CO). NuPAGE Novex
10% Bis- Tris gels, Platinum SYBR Green qPCR SuperMix-UDG w/ROX, were
purchased from Invitrogen (Carlsbad, CA). lRDye infiared secondary antibodies were

purchased from Invitrogen, Molecular probes (Eugene, OR).

3.2. Rat Model of Type 1 Diabetic Retinopathy

Male Sprague-Dawley rats weighing 237-283 grams were made diabetic with a single
intraperitoneal injection of 65 mg Streptozotocin (STZ) (Sigma, St. Louis, MO) per kg
body weight. STZ is a pancreatic B-cell cytotoxin, similar to glucose, transported into the
cell by the glucose transport protein GLUT2. Starting with day 7 after STZ injection,
Neutral Protamine Hagedom (NPH) insulin injections (with a dose of 0-3 units/day) was
administrated in order to achieve slow weight gain, but allowing hyperglycemia in the
range of 20mM blood glucose. The weight gain and blood glucose measurements from all
the groups studied are presented in Table 4.1 and Table 4.2. The retina was harvested and

analyzed at several time points (one month and nine months of diabetes).

163

3.3. The Experimental Diet

In our study, we maintained the recommended 10% caloric intake from lipid and replaced
5% of this value with fish oil in the treatment group. Two dietary experimental groups of
animals were established in both control non-diabetic and diabetic animals: a fish oil diet
(co-3-PUFA) and a vegetable oil diet group. The diets were purchased from Dyets Inc.
(Bethlehem, PA) and are based on AIN-93M purified rodent diet composition with 10%
caloric intake as soybean oil (standard rodent diet ingredient contains 50.8% linoleic acid)
or 5% as soybean oil and 5% as Menhaden oil (oil from a small plankton eating fish
containing 10.26% DHA and 14.16% EPA). The n6/n3 fatty acid ratio in vegetable oil
diet is 8/1, reflecting 15-16.7/1 ratio found in most Western diets (25). The n6/n3 ratio in
fish oil diet represents 1.5/1, reflecting the ratios associated with prevention of
cardiovascular disease (26) or suppression of inflammation in patients with rheumatoid
arthritis (27). The fatty acid composition of the control, standard diet and fish oil (DHA)-

enriched diet are presented in Table 4.3.

3.4. Isolation of Retinal Vasculature

Rat retinal vasculatures were isolated by trypsin digestion method as previously described
(28-29). The eyes were fixed in formalin for two weeks. The day before trypsin digest,
the eyes were opened coronally, cornea and lens were discarded and posterior globe was
placed into distilled water. The retina was loosened from sclera with metal micro-spatula
and cut loose at Optic nerve with microscissors in the way that optic nerve head remained

attached to the retina. The retina was placed in a perforated sample holder under gently

164

running tap water overnight to remove formalin fixative. The next morning the retina was
placed in 25 ml glass beaker with 5 ml of 3% crude trypsin in Tris HCl buffer, pH 7.8
with 0.2M NaF and digested for about 90 min. The inner limiting membrane was
removed and the digested retina was placed into a Petri dish containing distilled water
using a glass pipette as a transfer tool. The retina was gently brushed with cat whisker
hair to dislodge neural elements. The vessel network was cleaned three times in pure
distilled water, transferred onto mounting slide, and stained with hematoxylin and

periodic acid-Schiff.

3.5. Histological Assessment of Retinal Capillaries

The mid-retina areas along each major vessel of the vascular bed were systematically
investigated (6-8 fields per retina) to quantify the acellular capillaries at 250x
magnification (~0.32 mm2 per field). A capillary was counted as acellular if it had no
nuclei from junction to junction and comprises at least 20% of the diameter of

surrounding capillaries.

3.6. Simultaneous Extraction of Proteins and RNA

The method was performed as previously described (30). Each retina was homogenized
in 1 mL of TRIzol reagent. Then 200 pl Chloroform was added, the sample was covered
tightly, shacked vigorously for 15 seconds and let it stay at room temperature for 2-3

minutes; then centrifuge at 13,000 g for 15 minutes at 4°C. Phases separation was

165

obtained: the aqueous phase (RNA), the red organic phase (protein), the interphase
(DNA). Any residual DNA was precipitated from the organic phase with 100% ethanol
(300 pl), followed by a 2000g centrifugation for 5 minutes at 4 °C. The organic phase
was resolved in 14 mL of wash buffer (31.25mmol/L Tris-HCI, pH 6.8, 0.1% SDS
(sodium dodecyl sulfate)). After centrifugation at 4000g for 5 minutes at 23 °C, the upper
phase and wash buffer was transferred to a 5KD MWCO Amicon -15 (Millipore)
ultrafiltration tube. The Amicon -15 (Millipore) ultrafiltration tube was centrifuged at
3000g for 20 minutes at 23 °C. The filtrate was discarded and the same wash step was
performed 3 times with 14 mL of new wash buffer added each time. The protein solution

(approx 250 pl retentate) was transferred to an eppendorf tube; 5 pl of protease inhibitor

cocktail was added. The sample was stored in -80 °C until analysis.

3.7. Quantitative Real Time-Polymerase Chain Reaction (qRT-PCR)

Total RNA was extracted from rat and mice retinas. Specific primers for each gene are
listed in Table 1. Primers were designed using IDT DNA PrimerQuest software
(Coralville, IA). First strand cDNA was synthesized using the SuperScript II RNase H-
Reverse Transcriptase (Invitrogen Carlsbad, CA). Synthesized cDNA was mixed with 2x
SYBR Green PCR Master Mix (Invitrogen Carlsbad, CA).and different sets of gene-
specific forward and reverse primers and subjected to real-time PCR quantification using
the ABI PRISM 7900 Sequence Detection System (Applied Biosystems). All reactions
were performed in triplicate. The relative amounts of mRNAs were calculated by using
the comparative CT method (User Bulletin #2, Applied Biosystems). Cyclophilin was

used as a control and all results were normalized to the abundance of cyclophilin mRNA.

166

3.8. Western Blot

Proteins were extracted from rat retinas using a simultaneous extraction of proteins and
RNA method as previously described. Proteins were resolved by NuPAGE Novex 10%
Bis-Tris gels, transferred to nitrocellulose membrane and immunobloted using
appropriate primary antibodies followed by secondary lRDye infrared secondary
antibodies (Invitrogen, Molecular probes, Eugene, OR). Irnmunoreactive bands were
visualized and quantified by Odyssey infrared imaging system (Ll-COR Biosciences,

Lincoln, NE).

3.9. Immunofluorescence Assay

Rat retinal samples were fixed in Zinc Fixative (BD Biosciences, San Jose, CA) for 48
hours, then washed in PBS. Then the samples were blocked 60 minutes with 3% normal
serum in 0.3% Triton-x100 in TBS. The primary antibody incubation was overnight at
4°C on rocker. The next day tissue was rinsed in two changes of TBS+Tween20 (TBST),
five min each on rocker at room temperature. Secondary fluorescent antibody incubation
was three hours at room temperature on rocker. Then tissue was rinsed in three changes of

TBST for 10 min each and mounted on slides and covered with a coverslip.

3.10. Rat Model of Retinal Ischemia-Reperfusion

Male Sprague-Dawley rats weighing 300 grams were used. All procedures involving the

animal models adhered to the ARVO statement for the Use of animals in Ophthalmic and

167

Vision research. Retinal ischemia-reperfusion was created by temporal increase in
intraocular pressure to 90 mm Hg as previously described (31). The DHA treated group
was on a DHA-enriched diet for 12 days and has also been supplemented via tail vein
with 500 nmol/kg DHA one day prior and 250 nmol/kg DHA (32) immediately prior to
induction of I/R. The control, no treatment group was on standard rodent diet and
subjected to saline solution injection via the tail vein. The retinas were isolated two days
after retinal I/R, the time point that corresponds to inflammatory changes similar to the

early stage of diabetic retinopathy.

3.11. Slide Preparation and Quantification of the ASMase Staining of the Blood

Vessels

Rat retinal samples were fixed in Zinc Fixative (BD Biosciences, San Jose, CA) for 48
hours, followed by routine automated vacuum infiltrating tissue processing to paraffin
block. Zinc fixed paraffin blocks were sectioned on a rotary microtome at 4 — 5 microns
and placed on 3-aminoalkylethoxysilane coated slides, dried overnight at 56°C. Slides
following de—paraffinization and hydration to distilled water were placed in TBST
(Scytek Labs — Logan, UT) to allow for pH adjustment to 7.4. All slides were stained on
a Dako Autostainer (Dako North America — Carpinteria, CA) with TBST rinses between
each staining reagent. Non-specific protein block utilized normal serum from the host of
the secondary antibody for 30 minutes (Vector Labs — Burlingarne, CA). Endogenous
biotin block in Avidin D (Vector) / d-Biotin (Sigma — St. Louis, MO) was used for 15

minutes. Primary antibodies for ASMase and caveolin-1 were applied at optimized

168

dilutions in normal antibody diluent (Scytek) for 1 hour and then biotinylated anti-Host
species IgG H+L diluted to l lug/m1 (Vector) in NAD was applied for 30 minutes. R.T.U.
Vectastain® Elite ABC Reagent (Vector) was used next for 30 minutes. The reaction was
completed with Vector Nova Red peroxidase chromogen (Vector) for— 15 minutes. Slides
were then removed from the auto-stainer; counterstained with Gill 2 Hernatoxylin,
differentiated, dehydrated, cleared and coverslipped with synthetic mounting medium.
The pictures were obtained using a Micropublisher 3.3 Megapixel Color Digital Camera
and the amount of staining was quantified using a MetaMorph imaging system

(Molecular Devices, Downingtown, PA)

3.12. Statistical Analysis

Data are expressed as the mean i SEM for gene expression measurements and mean i SD
for western blot quantification and acellular capillary quantification. Factorial ANOVA
with post hoc Tukey test (GraphPad PrismS, GraphPad Software, San Diego, CA) was
used for comparing the data obtained from multiple independent samples. T-test was used

to compare data obtained from two independent measurements. Significance was

established at P<0.05.

169

4. Results

4.1. DHA Enriched Diet Prevents the Upregulation of Retinal ASMase in One

Month Diabetic Rat Model

The natural history of diabetic retinopathy has been divided into a very early pro-
inflarnmatory stage and a later, proliferative stage. We assessed the very early stage of
diabetic retinopathy in rats affected by type 1 diabetes for a one month period of time
(Figure 4.1, 4.2 and 4.3). To investigate whether DHA has a beneficial effect on DR,
diabetic rats were subjected to either a soybean oil diet (standard rodent diet) or fish oil
(DHA) enriched diet for the duration of the study. The control rats were fed a soybean oil
diet. As activation of SMases is an early critical step in inflammatory signaling, we first
determined the effect of diabetes on SMases. Interestingly, we found that ASMase gene
expression (Figure 4.1) and protein level (Figure 4.3 A,B) was dramatically increased in
retinas isolated from diabetic rats on soybean oil diet compared to control, healthy rats on
the same diet. Moreover, a DHA (fish oi1)-enriched diet prevented the upregulation of
ASMase gene expression (Figure 4.1) and protein levels (Figure 4.3 A,B) in the diabetic
rat retina. We found no effect of diabetes or the DHA enriched diet on NSMase gene
expression (Figure 4.1), strongly suggesting that ASMase rather than NSMase might play
a more important role in mediating vascular and retinal inflammation in the early stage of

diabetic retinopathy.

170

4.2. DHA Enriched Diet Prevents the Upregulation of Retinal
Proinflammatory/Angiogenic Mediators in One Month Diabetic Rat Model

At this initial stage of DR we also found a dramatic upregulation of
proinflammatory and pro-angiogenic mediator gene expression, as interleukin 1B (IL-1 [3),
IL-6 and vascular endothelial growth factor (VEGF) gene expression was increased in the
retinas of diabetic rats fed a soybean oil diet (Figure 4.2). VEGF protein level (Figure 4.3
A,C) was also increased two fold in the diabetic retina from rats fed a soybean oil diet
compared to control retina from rats on the same diet. Intercellular adhesion molecule 1
(ICAM-l) plays a critical role in mediating leukocyte adherence to endothelial walls,
leukostasis, and endothelial cell injury and death. ICAM-l gene expression (Figure 4.2)
was increased three fold in the retinas of diabetic rats fed a soybean oil diet when
compared to non—diabetic controls. More importantly, rats fed a DHA (fish oil)- enriched
diet did not manifest the diabetes induced upregulation of ICAM-1, VEGF, IL-lB and IL-
6 gene expression (Figure 4.2), as well as VEGF protein level (Figure 4.3 A,C), strongly
demonstrating a DHA protective effect against the development of early inflammation in

the diabetic retina.

4.3. DHA Enriched Diet Prevents the Upregulation of Retinal ASMase in Nine

Months Diabetic Rat Model

We next investigated the effect of long-tenn diabetes and a DHA-enriched diet on
sphingomyelinase expression in the retina. We found significant upregulation of retinal

ASMase gene expression nine months, after induction of diabetes in rats. Interestingly,

171

there was no change in NSMase gene expression between control rats fed a soybean oil
diet and diabetic rats fed either a soybean oil or DHA rich diet (Figure 4.4). These results
suggest that ASMase rather than NSMase may play the more important role in mediating
retinal inflammation and dysfunction in nine-month old diabetic rats. Moreover, a DHA-
enriched diet significantly prevented diabetes-induced upregulation of ASMase gene

expression (Figure 4.4) and protein levels (Figure 4.5 A,B).

4.4. DHA Enriched Diet Prevents the Upregulation of Retinal

Proinflammatory/Angiogenic Mediators in Nine Months Diabetic Rat Model

Pro-inflammatory and pro-angiogenic molecules, IL-IB and VEGF, were
significantly increased in the nine month-old diabetic rat retinas. Feeding diabetic rats a
DHA enriched diet prevented upregulation of IL-lB and VEGF gene expression (Figure
4.6). We found no change in ICAM-l gene expression between the retinas from nine
month-old diabetic rats and age matched controls (Figure 4.6). The lack of difference
between control and diabetic groups was due to an increase in ICAM-l gene expression

with age in control rats, rather than a decrease in ICAM-l level in diabetic animals

4.5. DHA Enriched Diet Prevents Retinal Vessel Loss in Nine Month-Old Diabetic

Rat Model

To determine whether a DHA enriched diet has a protective effect against the

diabetes induced retinal vessel loss, a hallmark for an advanced stage of retinopathy, rat

172

retina vasculature was isolated and the number of degenerated, acellular capillaries was
assessed in the retinas of nine month-old diabetic rats subjected to either a DHA-enriched
diet or a standard rodent, control diet. Wefound that the number of degenerate, acellular
capillaries was dramatically increased in diabetic retinas as compared to control retinas
isolated from rats fed soybean oil diet (Figure 4.7 A,B). Moreover, retinas from diabetic
rats on a DHA-enriched diet showed significantly fewer acellular capillaries then retinas
from diabetic rats fed a soybean oil diet (Figure 4.7 A,B). These data suggest that DHA
protects the retinal vasculature against vessel loss, a marker of advanced diabetic

retinopathy.

4.6. Upregulation of ASMase and Proinflammatory Cytokines TNFa and lL-lfl in
Human Diabetic Retina

Next we corroborated our findings in diabetic rats with studies in human diabetic
retinas and showed that retinas from diabetic donors exhibited increased ASMase, TNFa

and lL-lB mRNA levels when compared to control donors (Figure 4.8).

4.7. ASMase Expression in the Rat Retinal Vasculature

Our cell culture data demonstrate that caveolar ASMase expression in endothelial
cells rather than other retinal cells plays a key role in cytokine-induced endothelial
activation. To determine the distribution pattern of ASMase in vivo, we performed
immunohistochemistry of retina flatmounts and cross-sections of the retinal layers in rats.

ASMase is an important lysosomal enzyme that is ubiquitously expressed in most tissues.

173

We found basal ASMase expression in all retinal layers (Figure 4.9). Retinal blood
vessels, however, exhibited the highest positive staining for ASMase (Figure 4.9) which

colocalized with caveolin-1 (Figure 4.9).

4.8. DHA Supplementation Prevents the Upregulation of Vascular ASMase in the
Rat Retina Injured by HR

Next, we investigated the role of ASMase in the activation and degeneration of
retinal vessels using a well established model of accelerated ischemia/reperfusion (I/R)
injury that closely resembles diabetic retinopathy (31). ASMase expression was examined
in the retinal tissue two days following the I/R injury. Irnmunohistochemistry analysis of
retinal cross-sections obtained from uninjured and HR injured rat eyes, showed that I/R
injury specifically increased ASMase expression in the retinal vasculature (Figure 4.10)
when compared to uninjured eyes, with little or no effect on ASMase expression in other
retinal cells (Figure 4.10). As DHA was an effective ASMase inhibitor in our cell culture
experiments, we next determined the effect of this w3-PUFA on ASMase expression in
the rat retinal tissue isolated from uninjured and I/R injured eyes. DHA supplementation
prior to retinal injury specifically prevented ASMase upregulation in the retinal

vasculature (Figure 4.10).

174

Table 4.1. Body weight gain and blood glucose of experimental animals with one month

of diabetes. Data are mean i SD.

 

 

 

 

 

Weight gain, g/day Blood Glucose (mM)
Control 8.20i1.38 4.8i0.85
Diabetic control diet 4.13:1.6 25:12.5
Diabetic DHA 4.05i1.79 21.7i13.12
supplemented diet

 

 

 

 

Table 4.2. Body weight, blood glucose and HbAlc of experimental animals with nine

month of diabetes. Data are mean i SD.

 

 

 

 

 

 

 

Weight at the end of Blood Glucose HbA 1c (%)
the study (g) (mM)
Control 733.5 i74.24 5.04i0.348 5.3i0.6301
Diabetic control diet 547.8i71.87 23.14i4.21 l6.228i2.649
Diabetic DHA 586.83i88.27 18.15i5.023 13.05i3.117
supplemented diet

 

 

 

175

Table 4.3. Fatty acid composition of the control diet and DHA (fish oil) - enriched diet

 

Fatty acid Control diet

(percent of total)

DHA supplemented diet

mercent of total)

 

 

 

 

20:5n3 0.00 9.49
18:3n3 7.16 6.46
18:3n6 0.00 0.00
22:6n3 0.00 8.90
16:1n7 0.00 6.73
20:4n6 0.00 0.00
22:5n3 0.00 0.00
18:2n6 45.68 21.88
20:3n6 0.00 0.00
16:0 18.98 26.19
18:1n9 22.88 19.40
18:0 5.31 0.95
n6/n3 ratio
Control diet 6.38
DHA supplemented diet 0.88

 

 

 

 

176

Table 4.4. RT-PCR primers for rat

 

 

Gene Sense sequence Anti-sense sequence
ASMase caactatgggctgaagaagga acagctgactggcacacatt
NSMase ttctcgagctttgtccagggttc caggaagtgcttctttggctggt
ICAM-l ccaccatcactgtgtattcgtt acggagcagcactactgaga
VEGF A gctctcttgggtgcactgg caccacttcatgggctttct

IL-l B caaggagagacaagcaacga gtttgggatccacactctcc

Cyclophilin cttcttgctggtcttgccattcct tggatggcaagcatgtggtctttg

 

Table 4.5. RT-PCR primers for human

 

 

Gene Sense sequence Anti-sense sequence
ASMase caacctcgggctgaagaa tccaccatgtcatcctcaaaa
TNFa tccttcagacacctcaacc aggccccagtttgaattctt
IL-1 [3 gggcctcaaggaaaagaatc ttctgcttgagaggtgctga
Cyclophilin caagactgagtggttggatgg tggtgatcttcttgctggtct

 

177

0.04 - *

 

 

 

:__c_ 0.035
E 0.03 --
_c_>

3, 0.025
b 0.02
.9

is 0.015
2

g 0.01
l—

o _;__.._ ____

ASMase NSMase

 

E1 Control soybean oil
I

l Diabetes soybean oil
Diabetes fish oil (DHA-enriched)

 

 

Figure 4.1. The effect of a DHA-enriched diet on ASMase retinal gene expression in
rats with one month of diabetes. Retinas from rats fed either a soybean oil (standard
rodent) diet or DHA (fish oil)-enriched diet, isolated one month afier induction of
diabetes were analyzed by RT-PCR for ASMase and NSMase gene expression. Gene
expression from retinas isolated from rats subjected to a soybean oil diet (control, white
bar; diabetic, black bar) and a DHA-enriched diet (diabetic, gray bar). The relative
amounts of mRNAs were calculated by using the comparative CT method. Cyclophilin
was used as a control and all results were normalized to the abundance of cyclophilin
mRNA. n= 5 to 8 animals; *P<0.05 compared to control soybean oil; #P<0.05 compared
to diabetes soybean oil.

178

0.0025 0.12

 

.5

= *

§ 0.002 0-1 '

'5 0.08 .

6‘ 0.0015

3 0.06 .

'5 0.001 .

g 0.04 J

m l

1': “-0005 0.02 —3
0 - 0 —

.5 0.05 -

E

2 0.04

9.

o 0.03

b

.9

a 0.02

(D

g

r: 0.01 ,-

o
IL-1B IL-6

Figure 4.2. The effect of a DHA-enriched diet on retinal inflammatory/angiogenic
molecule gene expression in rats with one month of diabetes. Retinas from rats fed
either a soybean oil (standard rodent) diet or DHA (fish oil)-enriched diet, isolated one
month after induction of diabetes were analyzed by RT-PCR for inflammatory/angiogenic
molecule expression. RT-PCR analysis of ICAM-1, VEGF, IL-lB and IL-6 from retinas
isolated from rats subjected to a soybean oil diet (control, white bar; diabetic, black bar)
and a DHA-rich diet (diabetic, gray bar). The relative amounts of mRNAs were
calculated by using the comparative CT method. Cyclophilin was used as a control and all
results were normalized to the abundance of cyclophilin mRNA. n= 5 to 8 animals;
*P<0.05 compared to control soybean oil; #P<0.05 compared to diabetes soybean oil.

179

 

Diabetes

supplemented
Control Diabetes with DHA

' u l , 1
VEGF ----i
Tubulin --‘—-

 

 

   
   

 

 

B C.
ASMase VEGF

1 f- * 1.5 7
.E ‘
:3, i
3 1 _,.
l: f
E '
9, T
9 0.5 .
°- ;

0 .;__

 

Figure 4.3. The effect of a DHA-enriched diet on retinal ASMase and VEGF protein
expression in rats with one month of diabetes. (A) Immunoblot of ASMase and VEGF
protein levels in retinas isolated from control rats fed a soybean oil diet (white bar) and
from diabetic rats fed a soybean oil diet (black bar) and DHA-enriched diet (gray bar).
Quantitative analysis of the data presented in panel (A) for (B) ASMase and (C) VEGF
expression. The results are meaniSD. (n= 4 rats in each group; *P<0.05 compared to
control soybean oil; #P<0.05 compared to diabetes soybean oil).

180

0.25

 

g . *
E 0.2 ~
0.
2
9.
o 0.15
b
.9
b 0.1
(I)
C
E?
l- 0.05 —
0 .______

 

ASMase NSMase

 

[J Control soybean oil
I Diabetes soybean oil

1 Diabetes fish oil (DHA-enriched)

 

Figure 4.4. The effect of a DHA-enriched diet on retinal ASMase gene expression in
rats with nine months of diabetes. Retinas from rats fed either a soybean oil (standard
rodent) diet or DHA (fish oil)-enriched diet, isolated nine months after induction of
diabetes were analyzed by RT-PCR for ASMase and NSMase gene expression. Gene
expression from retinas isolated from rats subjected to a soybean oil diet (control, white
bar; diabetic, black bar) and a DHA-rich diet (diabetic, gray bar). The relative amounts
of mRNAs were calculated by using the comparative CT method. Cyclophilin was used
as a control and all results were normalized to the abundance of cyclophilin mRNA. n= 5
to 8 animals; *P<0.05 compared to control soybean oil; #P<0.05 compared to diabetes
soybean oil.

181

A. Diabetes

supplemented with
Control Diabetes DH A

I 11 ll

 

    
  

 

ASMase""" - , “'-
Tubulin —— “
B.
ASMase

Protein/Tubulin

 

 

 

’ El Control soybean oil

I I Diabetes soybean oil
E E3 Diabetes fish oil (DHA-enriched)
L

 

 

Figure 4.5. The effect of a DHA-enriched diet on retinal ASMase protein level in
rats with nine months of diabetes. (A) Immunoblot of ASMase protein levels in retinas
isolated from control rats on standard rodent diet (soybean oil diet) (white bar) and
diabetic rats on soybean oil diet (black bar) or DHA-enriched diet (gray bar). Equal
amounts of protein were added to each lane, as confirmed by tubulin levels. (B)
Quantitative analysis of the data presented in panel (A) for ASMase expression. The
results are meaniSD. (n=4 rats in each group. *P<0.05 compared to control soybean oil;
#P<0.05 compared to diabetes soybean oil).

182

0.0025 012 . 0.0006

   
  

 

c , *
% 0.0004 -~;
5 0.0015 - g
b 0.0003 7
.3 0.001 I
3 0.0002 s
C ‘ I
(0 0.0005 - .:
1': 0.0001 &
0 ~ 0 -

 

lL-1B

 

ED Control soybean oil

I Diabetes soybean oil
1531 Diabetes fish oil (DHA-enriched)

 

 

 

Figure 4.6. The effect of a DHA-enriched diet on retinal inflammatory/angiogenic
molecules gene expression in rats with nine months of diabetes. Retinas from rats fed
either a soybean oil (standard rodent) diet or DHA (fish oil)-enriched diet, isolated nine
months after induction of diabetes were analyzed by RT-PCR for
inflammatory/angiogenic molecule expression. RT-PCR analysis of ICAM-1, VEGF and
lL-lB from retinas isolated from rats subjected to a soybean oil diet (control, white bar;
diabetic, black bar) and a DHA-rich diet (diabetic, gray bar). The relative amounts of
mRNAs were calculated by using the comparative CT method. Cyclophilin was used as a
control and all results were normalized to the abundance of cyclophilin mRNA. n= 5 to 8
animals; *P<0.05 compared to control; #P<0.05 compared to diabetes soybean oil).

183

Control Diabetes

 

Soybean oil Soybean oil Fish oil

5
‘
’ .

(control dis!)

0, if:
x f *:

‘ ...". 1 I
fi ‘15 ‘r-‘\. ‘1
x ‘, .5 \ . 5
fi$ .- ‘ v. \
., 10,. I 3‘ _‘i 1' \\. - V.
‘ bdhfl' \. R“ t

 

Figure 4.7. The effect of a DHA-enriched diet on retinal vessel loss in rats with nine
months of diabetes. Retinas from rats fed either a soybean oil (standard rodent) diet or
DHA (fish oil) enriched diet, isolated nine months after induction of diabetes were
analyzed by microscopy of mounted retinal vasculature to visualize acellular capillaries.
(A) Acellular (degenerated) capillary formation (black arrows) in the retinal vasculature
isolated from control and nine months diabetic rats fed a soybean oil diet (n= 8) or a DHA
enriched diet (n= 8).

184

 

16

NE 14

E

g 12 -

£3 10 —~

% a -

0

LB 6 ~

2

6 4

O

< 2 _
0

 

E1 Control soybean oil E
. I Diabetes soybean oil E

i
E [1 Diabetes fish oil (DHA-enriched) ‘

Figure 4.7 (continued). The effect of a DHA-enriched diet on retinal vessel loss in
rats with nine months of diabetes. (B) Quantification of the total number of acellular
capillaries (average per field, 11: 8 animals per group). At least eight fields of retina were
counted in duplicates by two independent investigators, *P<0.05 compared to control;
#P<0.05 compared to diabetes soybean oil red bar = 550nm; black bar = 1 10m.

185

   

 

ASMase TNFa
C 0.03 — . 0.0015 —
=72 ..
3
3. 0.02 0.001
b
.g .
§ 0.01 0.0005
g s

o n..___ 0 j __ _
C.
IL-1B

5 0.004

a ..

i 0.003 f

0 ;

E1 0.002 ‘j

5 0.001

)—

 

 

Figure 4.8. Changes in retinal gene expression in control and diabetic human retina.
RT-PCR analysis of (A) ASMase, (B) TNFa and (C) IL-lB from retinas isolated from
postmortem human control (white bars) and diabetic eyes (black bars). The relative
amounts of mRNAs were calculated by using the comparative CT method. Cyclophilin
was used as a control and all results were normalized to the abundance of cyclophilin
mRNA. *P<0.05 compared to control.

186

 

ASMase Caveolin-1 Mer ed

 

 

 

 

 

 

 

 

 

 

 

Figure 4.9. Rat retinal ASMase expression. (A) Colocalization of ASMase and
Caveolin-1 in normal rat retina flat mount. (B) ASMase expression in normal rat retina
showing ASMase positive staining of retinal vessels (black arrow) and perinuclear
lysosomal compartment of retinal ganglion cells (black arrow head). (C) Caveolin-1
expression in normal rat retina showing caveolin-1 positive staining of retinal vessels
(black arrow); red bar = 55 um.

187

ASMase staining
Uniniured eye I I l/R iniured eye I

11~-.,.-..,.;; II ‘ I
$ . ”filyJ'I ‘ ‘3 ’ . .

 

 

 

Secondary antibody control ,
Uninjured eye I I l/R injured eye I
‘ * ~ ~ ' he :1-

i

.

5‘5 0.. ......f ‘3‘ 0"”. " ' "J.,
. ‘ v .0

. I

       

          

8'. o

 

Figure 4.10. Retinal vascular ASMase expression after retinal I/R injury in rats. (A)
Specific increase in ASMase expression in the rat retinal vessels (black arrow) from I/R

injured eye (A, right) compared to control (A, left). (B) Secondary antibody control for
(A).

188

C: ASMase staining

 

 

I Uninjured eye I I l/R injured eye I

 

  

a“ €8.43.
- am}

     
  

' .5 .1 .xi
”min! f 9

.‘3f2’1'v’

 

 

 

 

 

 

 

 

 

 

Figure 4.10 (continued). Retinal vascular ASMase expression after retinal I/R injury
in rats. DHA supplementation prevents the increase in vascular ASM expression in UK
injured eye (C, right). (D) Secondary antibody control for (C).

189

Retinal vessels

 

     

 

 

 

 

., 0.025 -— 11
C .
D
gag 0.02
§ ‘61 ,
g 2 0.015 ~
on .9 ..
U) .
gfi 0.01
‘2 “"273
E 0.005 «3
8
0
Control DHA supplementation
F.
Neuroretina
1:3 0.025 -
3 1
on 2
.§% 0.02 r
E w
w '8 0.015 ‘3
a -
9% 0.01 -
‘2 a
g 0.005 -«
C
0

 

 

 

 

 

Control DHA supplementation

Figure 4.10 (continued). Retinal vascular ASMase expression after retinal I/R injury
in rats. Quantification of ASMase staining in retinal vessels (E) and neuroretina (F). The
data is presented from five animals in each group, 11-17 retinal vessels per group;
*P<0.05 compared to all other groups. Red bar = 55 um, black bar = 220m.

190

5. Discussion

Early stage diabetic retinopathy is regarded as a low-grade chronic inflammatory
disease. Inflammation is part of the body's normal response to infection and injury;
however, when it happens in an uncontrolled or inappropriate manner it can induce
excessive damage to the host tissue and disease may ensue. Such uncontrolled or
inappropriate inflammatory responses also occur in diabetic retina and are characterized
by increased expression of adhesion molecules on the vascular endothelium and
leukocyte surfaces, sequestration of leukocytes to sites where they are not commonly
found, production of inflammatory mediators and ultimately damage to the host tissue
(2). The molecular mechanisms that lead to a persistent inflammation in the diabetic
retina are not well defined, but it seems that hyperglycemia and dyslipidemia, two
metabolical characteristics of diabetes mellitus, play an important role in the

inflammatory process.

Various biochemical pathways have been proposed to represent the link between
hyperglycemia and diabetic microangiopathy. Among these pathways, polyol
accumulation (33-34), oxidative stress and formation of reactive oxygen species (ROS)
(35), increase activity of various PKC isoforms (36-37), and glycation of key proteins
that lead to formation of advanced glycation end products (AGES) (38-40) are metabolic
pathways that have received much attention. Numerous studies have also demonstrated
that AGES production(41), PKCS activation(42), and ROS generation(43), have the
ability to activate sphingomyelinases. ASMase and NSMase are considered major
candidates for mediating stress-induced cell injury and death; however, these enzymes

have individual requirements for their activation and regulation. ASMase functions best

191

at a pH of 4.5-5.5, whereas optimal NSMase activation occurs at pH 7.4, and requires
magnesium or manganese (44). Diabetic hyperglycemia and hypoxemia induced local
retinal acidosis along with diabetes-related hypomagnesaemia (45) and this may
downplay the role of NSMase in diabetic retinopathy. Indeed, our study demonstrates that
ASMase rather than NSMase could represent a key player in mediating vascular damage

in diabetic retinopathy.

Dyslipidemia associated with diabetes has also been implicated in promoting the
inflammatory status in the retina. Along with DCCT/EDIC findings for a strong
association between dyslipidemia and diabetic retinopathy, our recent study has
demonstrated a significant decrease in total retinal (03 PUFAS, and especially DHA, that
was tightly coupled with inflammatory status in the diabetic retina (20). At the initial
stage of diabetic retinopathy, we also found in this study a dramatic increase in retinal
gene expression of the pro-inflammatory and pro-angiogenic mediators, IL-10 and VEGF
(Figure 4.2 and Figure 4.3 A,C). The involvement of IL-IB in pathogenesis of diabetic
retinopathy has been recently defined in diabetic mice (46). Inhibition of caspase-1, the
enzyme involved in production of IL-10 cytokine or inhibition of IL-10 signaling using
IL-1|3 receptor deficient mice protected the mice from development of retinal pathology
over seven months duration of diabetes (46). VEGF is a key molecule that plays a well-
recognized role in promoting vascular permeability (47) and pathological ocular
neovascularization (48). VEGF expression is mainly regulated by hypoxic stimulus; yet
there is evidence that it also accumulates in the retina early in diabetes, before any retinal

hypoxia is present (49-51) and our results are in accordance with these studies.

192

Inflammatory mediators can promote the activation and degeneration of capillary
endothelial cells with subsequent profound impact on vascular permeability and viability.
ICAM-l plays a critical role in mediating leukocytes adherence to endothelial walls,
leukostasis, increased movement of leukocytes from the bloodstream into the surrounding
tissue and endothelial cell injury and death. Pro-inflammatory cytokines, including IL-10
and VEGF have a pronounced effect on upregulation of ICAM-1 expression in the retina.
We found in our study that ICAM-l gene expression was increased three fold in the

retina from diabetic rats on a standard diet when compared to control (Figure 4.2).

ASMase is regarded as an initial pivotal responder in diverse cellular processes of
physiologic and pathologic importance. ASMase induction by various cytokine and stress
stimuli has been reported to play an essential role in cytokine-mediated inflammation,
apoptosis, cellular differentiation and cellular senescence (52-53). At the initial stage of DR,
we show that ASMase gene expression and protein levels were significantly increased in the
diabetic retina. The increased expression of many inflammatory proteins is regulated at the
gene transcription level through the activation of inflammatory transcription factors,
including nuclear factor kappa B (N F-KB), activator protein 1 (AP-1), specificity protein 1
(SP1) and other members of the nuclear receptor family (54-55). Analysis of the ASMase
promoter region identified several promoter elements essential for basal and inducible
ASMase gene expression; among these, SP-l, AP-l and AP-2 are mostly characterized and a
concerted action of these transcription factors appears to be essential in increased ASMase
expression (56-57). Interestingly, the upregulation of ASMase expression in diabetic retina
could potentiate the cytokine pleiotropic signaling properties in retinal inflammation.

Moreover, I/R injury-mediated upregulation of ASMase expression specifically in the retinal

193

vasculature may have a profound effect on cytokine induced inflammatory process in the
vascular endothelium, which ultimately results in endothelial cells activation, dysfunction

and death.

Omega-3 PUFAS, and particularly DHA are regarded as potent anti-inflammatory
agents and they may be of therapeutic benefit in a diversity of acute and chronic
inflammatory conditions (58). Our findings support the anti-inflammatory effect of DHA
and show that DHA supplementation decreases the expression of inflammatory mediators
and ICAM-1 in diabetic retina. More importantly, DHA decreases ASMase expression in the
diabetic retina and I/R injured vasculature, and this in vivo study recapitulates our
observations in in vitro settings where ASMase downregulation by DHA or ASMase gene
silencing had significant inhibitory effects on cytokine-induced upregulation of adhesion
molecules expression in retinal endothelium. Furthermore, DHA supplementation conferred
significant protection against the development of acellular capillaries, the hallmark of late

stage diabetic retinopathy.

In conclusion, we established the protective effect of DHA in in vivo models of
diabetic retinopathy. DHA downregulates ASMase expression, prevents retinal
inflammation and retinal capillary loss in rats with DR. Moreover, DHA modulates
vascular ASMase expression in the l/R injured rat retina. These results demonstrate that
ASMase may play a pivotal role in retinal inflammation and vascular degeneration in type
1 diabetic rat animal model. Consequently, in the following chapter, we directly address the
role of ASMase in retinal vascular microangiopathy using in vivo accelerated models of
retinopathy, including retinal ischemia-reperfusion and oxygen-induced retinopathy that

were induced in an ASMase deficient (ASMase 4') mouse.

194

6. References

l.

10.

ll.

Sinclair, A.J., Bayer, A.J., Girling, A.J., and Woodhouse, K.W. 2000. Older
adults, diabetes mellitus and visual acuity: a community-based case-control study.
Age Ageing 29:335-339.

Kern, TS. 2007. Contributions of inflammatory processes to the development of
the early stages of diabetic retinopathy. Exp Diabetes Res 2007:95103.

Lee, P., Wang, CC, and Adamis, AP 1998. Ocular neovascularization: an
epidemiologic review. Surv Ophthalmol 43:245-269.

Meleth, A.D., Agron, E., Chan, C.C., Reed, G.F., Arora, K., Bymes, G., Csaky,
K.G., Ferris, F.L., 3rd, and Chew, E.Y. 2005. Serum inflammatory markers in
diabetic retinopathy. Invest Ophthalmol Vis Sci 46:4295-4301.

Joussen, A.M., Murata, T., Tsujikawa, A., Kirchhof, B., Bursell, S.E., and
Adamis, AP. 2001. Leukocyte-mediated endothelial cell injury and death in the
diabetic retina. Am J Pathol 158:147-152.

McLeod, D.S., Lefer, D.J., Merges, C., and Lutty, GA. 1995. Enhanced
expression of intracellular adhesion molecule-1 and P-selectin in the diabetic
human retina and choroid. Am J Pathol 147:642-653.

Limb, G.A., Chignell, A.H., Green, W., LeRoy, F., and Dumonde, DC. 1996.
Distribution of TNF alpha and its reactive vascular adhesion molecules in
fibrovascular membranes of proliferative diabetic retinopathy. Br J Ophthalmol
80:168-173.

Joussen, A.M., Poulaki, V., Le, M.L., Koizumi, K., Esser, C., Janicki, H.,
Schraermeyer, U., Kociok, N., Fauser, S., Kirchhof, B., et a1. 2004. A central role

for inflammation in the pathogenesis of diabetic retinopathy. Faseb J 18:1450-
1452.

Murata, T., Nakagawa, K., Khalil, A., Ishibashi, T., lnomata, H., and Sueishi, K.
1996. The relation between expression of vascular endothelial growth factor and
breakdown of the blood-retinal barrier in diabetic rat retinas. Lab Invest 74:819-
825.

Adamis, A.P., Miller, J .W., Bemal, M.T., D'Amico, D.J., Folkman, J ., Yeo, T.K.,
and Yeo, K.T. 1994. Increased vascular endothelial growth factor levels in the

vitreous of eyes with proliferative diabetic retinopathy. Am J Ophthalmol
118:445-450.

Gustavsson, C., Agardh, C.D., Hagert, P., and Agardh, E. 2008. Inflammatory

markers in nondiabetic and diabetic rat retinas exposed to ischemia followed by
reperfusion. Retina 28:645-652.

195

 

12.

l3.

14.

15.

l6.

17.

18.

19.

20.

21.

Liu, P., and Anderson, R.G. 1995. Compartmentalized production of ceramide at
the cell surface. J Biol Chem 270:27179-27185.

Mathias, S., Younes, A., Kan, C.C., Orlow, 1., Joseph, C., and Kolesnick, RN.
1993. Activation of the sphingomyelin signaling pathway in intact EL4 cells and
in a cell-free system by IL-1 beta. Science 259:519-522.

Yang, Z., Costanzo, M., Golde, D.W., and Kolesnick, RN. 1993. Tumor necrosis
factor activation of the sphingomyelin pathway signals nuclear factor kappa B
translocation in intact HL-60 cells. J Biol Chem 268:20520-20523.

Marathe, S., Schissel, S.L., Yellin, M.J., Beatini, N., Mintzer, R., Williams, K.J.,
and Tabas, I. 1998. Human vascular endothelial cells are a rich and regulatable

source of secretory sphingomyelinase. Implications for early atherogenesis and
ceramide-mediated cell signaling. J Biol Chem 273 :4081-4088.

K0, Y.G., Lee, J.S., Kang, Y.S., Ahn, J.H., and Seo, J.S. 1999. TNF-alpha-
mediated apoptosis is initiated in caveolae-like domains. J Immunol 162:7217-
7223.

Martin, M.U., and Wesche, H. 2002. Summary and comparison of the signaling
mechanisms of the Toll/interleukin-l receptor family. Biochim Biophys Acta
1592:265-280.

Wong, M.L., Xie, B., Beatini, N., Phu, P., Marathe, S., Johns, A., Gold, P.W.,
Hirsch, E., Williams, K.J., Licinio, J ., et al. 2000. Acute systemic inflammation
up-regulates secretory sphingomyelinase in vivo: 3 possible link between
inflammatory cytokines and atherogenesis. Proc Natl Acad Sci U S A 97:8681-
8686. .

Busik, J .V., Mohr, S., and Grant, MB. 2008. Hyperglycemia-induced reactive
oxygen species toxicity to endothelial cells is dependent on paracrine mediators.
Diabetes 57:1952-1965.

Tikhonenko, M., Lydic, T.A., Wang, Y., Chen, W., Opreanu, M., Sochacki, A.,
McSorley, K.M., Renis, R.L., Kern, T., Jump, D.B., et al. 2009. Remodeling of
Retinal Fatty Acids in an Animal Model of Diabetes: a Decrease in Long Chain
Polyunsaturated Fatty Acids is Associated with a Decrease in Fatty Acid
Elongases Elov12 and Elovl4. Diabetes.

Connor, K.M., SanGiovanni, J.P., Lofqvist, C., Aderman, C.M., Chen, J.,
Higuchi, A., Hong, S., Pravda, E.A., Majchrzak, S., Carper, D., et al. 2007.
Increased dietary intake of omega-3-polyunsaturated fatty acids reduces
pathological retinal angiogenesis. Nat Med 13:868-873.

196

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

Chen, W., Esselman, W.J., Jump, DB, and Busik, J .V. 2005. Anti-inflammatory
effect of docosahexaenoic acid on cytokine-induced adhesion molecule

expression in human retina] vascular endothelial cells. Invest Ophthalmol Vis Sci
46:4342-4347.

Opreanu, M., Lydic, T.A., Reid, G.E., McSorley, K.M., Esselman, W.J., and
Busik, J. 2010. Docosahexaenoic Acid Inhibits Cytokine Signaling in Human
Retinal Endothelial Cells by Downregulating Sphingomyelinases. Invest
Ophthalmol Vis Sci.

Futterman, S., and Kupfer, C. 1968. The fatty acid composition of the retinal
vasculature of normal and diabetic human eyes. Invest Ophthalmol 7: 1 05-108.
Simopoulos, AP. 2003. Importance of the ratio of omega-6/omega-3 essential
fatty acids: evolutionary aspects. World Rev Nutr Diet 92:1-22.

de Lorgeril, M., and Salen, P. 2003. Dietary prevention of coronary heart disease:
focus on omega-6/omega-3 essential fatty acid balance. World Rev Nutr Diet
92:57-73.

Cleland, L.G., James, M.J., and Proudrnan, SM. 2003. Omega-6/omega-3 fatty
acids and arthritis. World Rev Nutr Diet 92:152-168.

Zheng, L., Szabo, C., and Kern, T .S. 2004. Poly(ADP-ribose) polymerase is
involved in the development of diabetic retinopathy via regulation of nuclear
factor-kappaB. Diabetes 53:2960—2967.

Kern, T.S., Tang, J., Mizutani, M., Kowluru, R.A., Nagaraj, R.H., Romeo, G.,
Podesta, F., and Lorenzi, M. 2000. Response of capillary cell death to
aminoguanidine predicts the development of retinopathy: comparison of diabetes
and galactosemia. Invest Ophthalmol Vis Sci 41 :3972-3978.

Agardh, E., Gustavsson, C., Hagert, P., Nilsson, M., and Agardh, CD. 2006.
Modifying a standard method allows simultaneous extraction of RNA and protein,

enabling detection of enzymes in the rat retina with low expressions and protein
levels. Metabolism 55:168-174.

Zheng, L., Gong, B., Hatala, D.A., and Kern, TS. 2007. Retinal ischemia and
reperfusion causes capillary degeneration: similarities to diabetes. Invest
Ophthalmol Vis Sci 48:361-367.

Huang, W.L., King, V.R., Curran, O.E., Dyall, S.C., Ward, R.B., Lal, N.,
Priestley, J .V., and Michael-Titus, AT. 2007. A combination of intravenous and
dietary docosahexaenoic acid significantly improves outcome after spinal cord
injury. Brain 13023004-3019.

197

 

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

Engerman, R.L., and Kern, TS. 1984. Experimental galactosemia produces
diabetic-like retinopathy. Diabetes 33 :97-1 00.

Frank, R.N., Keim, R.J., Kennedy, A., and Frank, K.W. 1983. Galactose-induced
retinal capillary basement membrane thickening: prevention by Sorbinil. Invest
Ophthalmol Vis Sci 24:1519-1524.

Giugliano, D., Ceriello, A., and Paolisso, G. 1996. Oxidative stress and diabetic
vascular complications. Diabetes Care 19:257-267.

Shiba, T., Inoguchi, T., Sportsman, J .R., Heath, W.F., Bursell, S., and King, G.L.
1993. Correlation of diacyl glycerol level and protein kinase C activity in rat retina
to retinal circulation. Am J Physiol 265:E783-793.

Ways, D.K., and Sheetz, M.J. 2000. The role of protein kinase C in the
development of the complications of diabetes. Vitam Horm 60: 149-193.

Brownlee, M. 1992. Glycation products and the pathogenesis of diabetic
complications. Diabetes Care 15: 1 835-1 843.

Brownlee, M., Vlassara, H., and Cerarni, A. 1984. Nonenzymatic glycosylation
and the pathogenesis of diabetic complications. Ann Intern Med 101 2527-53 7.

Wautier, J .L., and Guillausseau, P.J. 2001. Advanced glycation end products,
their receptors and diabetic angiopathy. Diabetes Metab 27:535-542.

Patschan, S., Chen, J ., Polotskaia, A., Mendelev, N., Cheng, J ., Patschan, D., and
Goligorsky, MS. 2008. Lipid mediators of autophagy in stress-induced premature
senescence of endothelial cells. Am J Physiol Heart Circ Physiol 294:H1119-
1129.

Zeidan, Y.H., Wu, B.X., Jenkins, R.W., Obeid, L.M., and Hannun, Y.A. 2008. A
novel role for protein kinase Cdelta-mediated phosphorylation of acid
sphingomyelinase in UV light-induced mitochondrial injury. F aseb J 22: 1 83-193.

Dumitru, C.A., Zhang, Y., Li, X., and Gulbins, E. 2007. Ceramide: a novel player
in reactive oxygen species-induced signaling? Antioxid Redox Signal 921535-
1540.

Marchesini, N., and Hannun, Y.A. 2004. Acid and neutral sphingomyelinases:
roles and mechanisms of regulation. Biochem Cell Biol 82:27-44.

Pham, P.C., Pham, P.M., Pham, S.V., Miller, J.M., and Pham, RT. 2007.

Hypomagnesemia in patients with type 2 diabetes. Clin J Am Soc Nephrol 2:366-
373.

198

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

Vincent, J .A., and Mohr, S. 2007. Inhibition of caspase-l/interleukin-lbeta
signaling prevents degeneration of retinal capillaries in diabetes and galactosemia.
Diabetes 56:224-230.

Senger, D.R., Connolly, D.T., Van de Water, L., Feder, J., and Dvorak, HF.
1990. Purification and NH2-terminal amino acid sequence of guinea pig tumor-
secreted vascular permeability factor. Cancer Res 50:1774-1778.

Ng, E.W., and Adamis, AP. 2005. Targeting angiogenesis, the underlying
disorder in neovascular age-related macular degeneration. Can J Ophthalmol
40:352-368.

Gerhardinger, C., Brown, L.F., Roy, S., Mizutani, M., Zucker, C.L., and Lorenzi,
M. 1998. Expression of vascular endothelial growth factor in the human retina
and in nonproliferative diabetic retinopathy. Am J Pathol 152:1453-1462.

Segawa, Y., Shirao, Y., Yamagishi, S., Higashide, T., Kobayashi, M., Katsuno,
K., lyobe, A., Harada, H., Sato, F., Miyata, H., et al. 1998. Upregulation of retinal
vascular endothelial growth factor mRNAs in spontaneously diabetic rats without
ophthalmoscopic retinopathy. A possible participation of advanced glycation end
products in the development of the early phase of diabetic retinopathy.
Ophthalmic Res 30:333-339.

Sone, H., Kawakami, Y., Okuda, Y., Sekine, Y., Honmura, S., Matsuo, K.,
Segawa, T., Suzuki, H., and Yamashita, K. 1997. Ocular vascular endothelial
growth factor levels in diabetic rats are elevated before observable retinal
proliferative changes. Diabetologia 40:726-730.

Hannun, Y.A. 1997. Sphingolipid second messengers: tumor suppressor lipids.
Adv Exp Med Biol 400A:305-312.

Kolesnick, RN, and Kronke, M. 1998. Regulation of ceramide production and
apoptosis. Annu Rev Physiol 60:643-665.

Rahman, I. 2002. Oxidative stress, transcription factors and chromatin
remodelling in lung inflammation. Biochem Pharmacol 64:93 5-942.

Winyard, P.G., and Blake, DR. 1997. Antioxidants, redox-regulated transcription
factors, and inflammation. Adv Pharmacol 38:403—421.

Langrnann, T., Buechler, C., Ries, S., Schaeffler, A., Aslanidis, C., Schuierer, M.,
Weller, M., Sandhoff, K., de Jong, P.J., and Schmitz, G. 1999. Transcription
factors Spl and AP-2 mediate induction of acid sphingomyelinase during
monocytic differentiation. J Lipid Res 40:870-880.

199

57.

58.

Schuchman, E.H., Levran, O., Pereira, L.V., and Desnick, R.J. 1992. Structural
organization and complete nucleotide sequence of the gene encoding human acid
sphingomyelinase (SMPDI). Genomics 12:197-205.

Calder, P.C. 2006. n-3 polyunsaturated fatty acids, inflammation, and
inflammatory diseases. Am J Clin Nutr 83:15058-15198.

200

Chapter V

Genetic Deficiency of Acid Sphingomyelinase Prevents Upregulation of
Inflammatory/Angiogenic Mediators, Vessel Loss and Neovascularization in

the Retina of Mice with Microangiopathy Lesions

I . Abstract

Acid sphingomyelinase (ASMase) is a key candidate for stress-mediated cellular
activation and death. We have previously shown that ASMase was upregulated in
diabetic rat retina and closely correlated with increased inflammatory/pro-angiogenic
molecule levels in the same tissue. More importantly, ASMase expression specifically
increased in the rat retinal vasculature injured by ischemia/reperfusion when compared to
the non-injured eye. To directly investigate the role of ASMase in endothelial cell
activation and death and development of retinal acellular capillaries, we used accelerated
models of retinopathy by inducing ischemia! reperfusion (I/R) injury and oxygen-induced
retinopathy (OIR) in wild type (ASMaseV) or ASMaseT mouse retina. We found that
genetic deficiency of ASMase significantly prevented the increase in inflammatory/pro-
angiogenic molecules expression in I/R injured retina compared to wild type. Moreover,
I/R injured ASMaseT mouse retinas were protected from both vascular degeneration and
neovascularization. These results demonstrate that ASMase deficiency has a protective
effect on ischemia-driven endothelial cell dysfunction and death, preventing consequent
development of degenerated capillaries, as well as neovessel formation in the injured

retinas.

201

2. Introduction

Sphingomyelin is an important sphingolipid component of all mammalian
membranes and consists of a ceramide backbone with a phosphorylcholine moiety.
ASMase represents a key regulatory enzyme in sphingolipid metabolism, catalyzing the
hydrolysis of sphingomyelin to the bioactive lipid ceramide. ASMase major role in cell
signaling is closely linked to its ability to reorganize the plasma membrane where
sphingomyelin hydrolysis into ceramide facilitates the formation and coalescence of
plasma membrane lipid microdomains. These ceramide enriched microdomains are
believed to increase the local protein density, including receptors that often require
oligomerization for their activation or facilitate protein-protein interaction (1). Receptor
oligomerization and adaptor proteins recruitment appear to be a requirement for TNFa

and IL-1 Signalosome organization and intracellular inflammatory signaling (2-3).

Until the middle of the last decade, interest in ASMase biological role was
restricted primarily to researchers studying Niernann-Pick disease, a recessively inherited
lysosomal storage disease due to ASMase deficiency. At the same time, ASMase
knockout mice were generated and become available for research use (4). The use of
these ASMase knockout mice opened new avenues of research since these animals were
initially found to be resistant to radiation (5). Subsequently, there has been an advancing
crescendo of studies that have begun to shed significant light on our understanding on
ASMase involvement in cell survival and pathophysiology of cancer, diabetes, sepsis,

cardiovascular, pulmonary and neurological diseases (6). However, the role of ASMase in

202

the pathology of diabetic retinal microangiopathy has not been address so far and

represents the purpose of this study.

Vascular endothelium represents a very rich source of ASMase (5, 7) and, as
presented in the previous chapter, retinal vasculature exhibits significant positive staining
for ASMase when compared to other retinal tissues. Elevated ASMase concentration in
the vasculature makes it an excellent candidate for locally mediation of inflammatory and
cell death, signaling pathways that would ultimately result in vascular degeneration.

In this study, we directly tested the hypothesis that ASMase genetic deficiency
may be protective against the endothelial activation, injury and death, preventing the
retinal vessels loss and subsequent pathological angiogenesis. We have modeled
retinopathy in an ASMase knockout (ASMase '/') mouse model of Niemann-Pick disease
by using retinal ischemia/reperfusion (I/R) and oxygen-induced retinopathy (OIR), both
representing accelerated models of retinopathy; while the first model permits us to
investigate the formation of acelullar capillaries, the last one allows us the assessment of

neovessels formation.

The ischemia-reperfusion model induces retinal capillary degeneration, is well
established and has close similarities to diabetes induced retinal vascular
microangiopathy (8). Ischemia and reperfusion lead to individual effects on tissues
related to metabolically disregulation induced by both hypoxia and reoxygenation. The
arrest of blood supply in the retina results in cell death via both necrosis and apoptosis.
Hypoxia leads to disruption of cellular energy metabolism and initiates a harmful cascade
of events (9); induction of cytokine production, generation of reactive oxygen species

(ROS), degradation of the anti-oxidant system and extracellular buildup of glutamate are

203

the main changes secondary to ischemia (9-14). The reoxygenation process, although
effective in decreasing the ischemic damage, promotes additional cell injury and
apoptosis (15) and the reoxygenation-induced cell damage is largely due to generation of
reactive oxygen species (ROS) (16), increased levels of extracellular neurotransmitters
and waste products that damage previously unharmed cells when being reoxidized (9, 17-

18).

Ocular angiogenesis in response to tissue ischemia is commonly seen in various
clinical conditions, including diabetic retinopathy, retinopathy of prematurity and retinal
vein occlusion (19-20). The oxygen-induced retinopathy model uses a basic premise by
which exposure to hyperoxia during early development of retinal vasculature results in
the reduction of normal retinal vascular development and the degeneration of newly
developed vessels. Removal of the hyperoxic stimulus renders a relative ischemic state in
the retina, since the retina is no longer immersed in the high oxygen setting and lacks its
normal vasculature. This ischemic situation sets off a rapid pathologic angiogenic process
with formation of neovascular tufts which protrude out of the normal retinal vascular

plexuses into the vitreal cavity (21 ).

In the previous chapter, we demonstrated that ASMase gene expression and
protein level are significantly upregulated in diabetic rat retinas that exhibited an
inflammatory status and microangiopathy lesions; importantly, ASMase was significantly
increased in the vascular rat retina with I/R injury. ASMase is considered to be a first
responder in mediating signaling processes via a specific receptor and genetic ASMase
knockout mice exhibit protection from stress-induced cell death in various organs and

systems (6, 22-23). Therefore, in this study we hypothesized that the ASMase genetic

204

deficiency will render protection against upregulation of proinflammatory/angiogenic
mediators, vessel loss and pathological angiogenesis in the retina injured by ischemia-

reperfusion and hyperoxic stimulus, respectively.

205

 

3. Materials and Methods

3.1. Reagents

NuPAGE Novex 10% Bis- Tris gels, Platinum SYBR Green qPCR SuperMix-UDG
w/ROX were purchased from Invitrogen (Carlsbad, CA). Lyophilized pancreatic elastase
was purchased from Calbiochem (San Diego, CA). RT-PCR primers from IDT DNA
Technologies (Coralville, IA) (Table 1) were used. ASM antibody was a generous gift
from Dr. Richard Kolesnick. High Pure PCR Template Preparation Kit (Roche Applied
Science, Mannheim, Germany) was used for DNA isolation. DNase, RNase free water,
dNTPs, MgCl, Taq DNA Polymerase, Trackit Cyan/Yellow Buffer, Tracklt lkb Plus

DNA ladder were purchase from Invitrogen, (Carlsbad, CA).

3.2 Genotyping Transgenic Mice by PCR and Southemblot

3.2.1. DNA isolation from mouse tail. 0.2 to 0.5 cm of the mouse tail was used for DNA
isolation using a High Pure PCR Template Preparation Kit (Roche Applied Science,
Mannheim, Germany). The mouse tail was added to a1.5 mL nuclease-free
microcentrifuge tube, along with 200 uL Tissue Lysis Buffer and 20 uL Proteinase K
(reconstituted). The mixture was incubated overnight at 55 °C until the tissue was
completely digested. Using a 1 mL disposable syringe without needle, the tail sample was
lysed by shearing process. Then 200 uL of Binding Buffer and 100 uL of isopropanol
were added to the tube, mixed well and subjected to centrifugation for 5 minutes at
13,000 x g. Then the liquid sample was pipetted into the upper buffer reservoir of the
High Filter Tube that was previously inserted into one Collection tube. The entire High

Filter Tube assembly was inserted into a standard table-top centrifuge and centrifuged 1

206

 

minute at 8,000 x g. After centrifugation, the Filter Tube was removed from the
Collection Tube; the flowthrough liquid and the Collection Tube were discarded. The
Filter Tube was combined with a new Collection Tube; 500 uL Inhibition Removal
Buffer was added to the upper reservoir of the Filter Tube and centrifuged 1 min at 8,000
x g. After centrifugation, the Filter Tube was again removed from the Collection Tube;
the flowthrough liquid and the Collection Tube were discarded. The Filter Tube was
combined with a new Collection Tube; 500 uL Wash Buffer was added to the upper
reservoir of the Filter Tube and centrifuged 1 min at 8,000 x g. This final step was
repeated twice. To elute the DNA, the Filter Tube was inserted into a clean, sterile 1.5
mL microcentrifuge tube; 200 uL prewarmed Elution Buffer was added to the upper
reservoir of the filter tube and centrifuged 1 min at 8,000 x g. The eluted DNA was stored
at -20 0C for further analysis.

3.2.2. PCR amplification of the DNA sample. The following reagents were added to an
RNase, DNase fi'ee microcentrifuge tube (a total volume of 25 uL): DNase, RNase fi‘ee
water 16 uL, 10x buffer 2.5 uL, dNTPs (10mM) 1 11L, MgCl (150mM) 1.5 11L, Primer
Mix for ASMase (0.1 mM) 1.5 uL, Taq DNA Polymerase (Invitrogen, (Carlsbad, CA))
0.5 uL, DNA 2 uL. The following primers for ASMase have been used: GGCTACCCGT
GATATTGC, AGCCGTGTCCTCTTCCTTAC, CGAGACTGTI‘GCCAGACATC. The
PCR mix underwent the following cycling conditions: 94 °C for 3 minutes, followed by a
sequence (94 0C for 15 seconds, 64 °C for 30 seconds and 68 0C for 90 seconds) that was
repeated 35 times and then 72 °C for 7 minutes.

3.2.3. Southemblot identification of ASMase mutant allele. The PCR product (25 pL)

was mixed to 2.5 11L of loading buffer (Trackit Cyan/Yellow Buffer, Invitrogen,

207

 

Carlsbad, CA). 1.5% Agarose in 50 mL TBE buffer, mixed with 5 uL of 10,000x SYBR
safe stain was used to cast the gel. The Tracklt lkb Plus DNA ladder was added to the
first well. 10 uL of samples to analyze was added to the remaining wells. The ASMase
mutant allele was identified at 523 bp, whereas the wild-type allele was localized at 269

bp.

3.3. Mice. All procedures involving the animal models adhered to the ARVO statement
for the Use of animals in Ophthalmic and Vision research. Mouse model of retinal
ischemia-reperfusion. Two months old male C57BL/6J wild type (ASM+/+) and
C57BL/6J ASMase knock-out (ASM'/') mice were anesthetized. The anterior chamber of
one eye was cannulated with a 30 gauge needle attached to a line infusing sterile normal
saline solution. The intraocular pressure (IOP) was maintained at 80-90 mmHg; this
pressure was determined by the height and the volume of the perfusion bag. This resulted
in retinal ischemia, as confirmed by whitening of the iris and loss of the red reflex. The
other eye of the same animal was used as a control. The duration of ischemia was 90
minutes. After ischemia, the needle was withdrawn, followed by IOP normalization and
reperfusion of the retinal vasculature was documented visually. The mice were sacrificed
at different times after reperfusion (two days for assessment of inflammatory mediators
and seven days for assessment of acellular, degenerated capillaries). Mouse model of
oxygen-induced retinopathy (OIR). In the neonatal mouse model of oxygen-induced
retinopathy, 7-day-old (P7) C57BL/6J wild type and C57BL/6J acid sphingomyelinase
deficient (ASMase'/') mice were placed with their nursing dams in a 75% oxygen

atmosphere for 5 days (P7 to P12). Upon return to normal air, these mice develop retinal

208

 

neovascularization, with peak development occurring 5 days after their return to
normoxia. Retinal neovascularization was assessed at P17 as previously described (21,

24).

3.4. Isolation of Retinal Vasculature

Mouse retinal vascular beds were isolated by elastase digestion method as previously
described (25). The fixed mouse retinas were isolated and rinsed in deionized water as
described above. The next day retina was incubated for 15 minutes in an agitating water
bath preheated to 37°, in 40 units/ml elastase in 100 mmol/l sodium phosphate buffer
with 150 mmol/l sodium chloride and 5.0 mmol/l ethylenediamenetetraacetic acid
(EDTA) at pH 6.5. The tissue was washed overnight in 100 mmol/l Tris-HCl (pH 8.5) at
room temperature and then transferred to deionized water for removal of the now-
loosened vitreous and digested neural elements by gentle agitation using the sides of
forceps and the sides and ends of very fine sable brushes as needed. Then the retinas were
mounted by flotation in HPLC-grade water over slides. Once air-dried in a dust-free
environment, the mounts of the retinal vasculature were stained using the periodic-acid-

Schiff reaction and hematoxylin counterstaining.

3.5. Histological Assessment of Retinal Capillaries

The mid-retina areas along each major vessel of the vascular bed were systematically
investigated (6-8 fields per retina) to quantify the acellular capillaries at 250x

magnification (~0.32 mm2 per field). A capillary was counted as acellular if it had no

209

nuclei from junction to junction and comprises at least 20% of the diameter of

surrounding capillaries.

3.6. Simultaneous Extraction of Proteins and RNA

The method was performed as described in reference (26). The method has been

described in detail in chapter IV.

3.7. Quantitative Real Time-Polymerase Chain Reaction (qRT-PCR)
Total RNA was extracted from rat and mice retinas. Specific primers for each gene are

listed in Table l. The method has been previously described in chaper II.

3.8. Statistical analysis

Data are expressed as the mean i SEM for gene expression measurements and mean i SD
for western blot quantification and acellular capillary quantification. Factorial ANOVA
with post hoc Tukey test (GraphPad PrismS, GraphPad Software, San Diego, CA) was
used for comparing the data obtained from multiple independent samples. T-test was used
to compare data obtained from two independent measurements. Significance was

established at P<0.05.

210

4. Results

4.1. Retinal I/R Induces Upregulation of ASMase Gene Expression in the ASMase”+

Retina

Retinas from control and ischemia/reperfusion (l/R) injured eye were isolated
from wild type (ASMase+/+) and ASMase genetic deficient (ASMase'f) mice 2 days after
induction of the retinal damage. Real time qPCR analysis of ASMase gene expression
showed a significant upregulation of ASMase mRNA in wild type retinas injured by I/R
(Figure 5.1). Conversely, NSMase gene expression was not significantly increased in the
wild type retinas with I/R lesions, but was dramatically elevated in the I/R injured retinas
isolated fi'om the ASMaseT mice (Figure 5.2). These results confirmed the role of

ASMase as a major candidate in mediating stress-induced cell death.

4.2. ASMase Genetic Deficiency Prevents Upregulation of Cellular Adhesion

Molecules in the HR Injured Retina

Two days after ischemia-reperfusion, a time point that corresponds to an early
stage of diabetic retinopathy, the gene expression of intercellular adhesion molecule 1
(ICAM-l) (Figure 5.3 A) and vascular cell adhesion molecule 1 (VCAM-l) (Figure 5.3
B) was significantly increased in the wild type retinas injured by UK. Conversely, ICAM-
1 and VCAM-l gene expression was significantly decreased in injured retinas isolated
form ASMase'f mice when compared to retinas isolated from ASMaseJrl+ I/R injured

eye.

211

 

4.3. ASMase Genetic Deficiency Prevents Upregulation of Inflammatory/

Angiogenic Mediators in the I/R Injured Retina

Similarly to adhesion molecules expression, the gene expression of IL-1 [3 (Figure
5.4 A), TNFa (Figure 5.4 B) and VEGF (Figure 5.4 C) was significantly increased in the
wild type retinas injured by I/R two days after ischemia-reperfusion. More importantly,
ASMaseT mice exhibit IL-lB, TNFa and VEGF gene expression (Figure 5.4 A,B,C) at a
normal level in injured retinas compared to the wild type. Therefore, ASMase genetic
deficiency emerges as greatly protective against the development of an inflammatory

state in the I/R injured retina.

4.4. ASMase Genetic Deficiency Prevents the Development of Acellular Capillaries

in the [IR Injured Retina

Examination of mouse retinal vasculature in wild-type mice seven days after I/R
injury, when capillary degeneration response is greatest (8), showed a dramatic increase
in the number of acellular capillaries as compared to control retinas (Figure 5.5 A, C).
Furthermore, I/R injured ASMase'/' mouse retinas exhibited significantly fewer acellular
capillaries compared to the wild type (Figure 5.5 B, C). These results clearly demonstrate
that ASMase deficiency has a protective effect on ischemia-driven endothelial cell

dysfunction and death, preventing consequent formation of acellular capillaries.

212

 

4.5. ASMase Genetic Deficiency Prevents Ocular Neovascularization in OIR Mouse

Model

The experiments conducted so far investigated the protective role of ASMase
deficiency on the development of vascular retinal nonproliferative changes similar to
those that develop in diabetic retinopathy. Owing to the lack of an animal model, that
properly reproduces the development of proliferative diabetic retinopathy; we next
focused our study on an experimental model of ischemic neovascularization involving
oxygen-induced retinopathy (OIR). The OIR model allows the assessment of abnormal
retinal angiogenesis, which is similar to aberrant new vessel formation in late
proliferative stages of diabetic retinopathy. The degree of neovascularization was
quantified by counting the number of endothelial cell nuclei present on the vitreal side of
the inner limiting membrane (ILM). We found the average number of nuclei extending

past the ILM per 6 um cross-section to be 12.1i3.4 SD in hyperoxia-treated ASMaseT

mouse eyes, compared to 49.5i12.5 SD in hyperoxia-treated wild type mouse eyes
(Figure 5.6). These findings demonstrate that ASMase deficiency prevents the

development of pathological angiogenesis in the OIR retina.

213

Table 5.1. RT-PCR mouse primers

 

Gene

Sense sequence

Anti-sense sequence

 

ASMase

NSMase

ICAM-l

VEGF A

11.40

VCAM-l

TNFa

Cyclophilin

aaccctggctaccgagtttaccaa

atgcacactacttcagaagcggga

acactatgtggactggcagtggtt

tcaccaaagccagcacataggag

aagggctgcttccaaacctttgac

cccaggtggaggtctactca

tctcatgcaccaccatcaagg

attcatgtgccagggtggtga

tggcctgggtcagattcaagatgt

aggcattgagcaccagtccactta

tgaggctcgattgttcagctgcta

ttacacgtctgcggatcttggaca

atactgcctgcctgaagctcttgt

ccagatggtcaaagggataca

accactctccctttgcagaac

ccgtttgtgggtccagca

 

214

 

ASMase

 

 

. DControl eye
- *
g 03 —1. I l/R eye
_5- .
2 e
g 0.6 —I
0 i
:1 i
e 0.4 —«i
0 ;
(I) i
g ;
l: 0.2 ‘i
i
0 _L_

ASMase +/+

Figure 5.1. Retinal gene expression of ASMase in the HR retinopathy model. RT-
PCR analysis of ASMase in I/R injured retinas isolated from ASM” mice 2 days
following retinal ischemia-reperfusion injury. The relative amounts of mRNAs were
calculated by using the comparative CT method. Cyclophilin was used as a control ail/(1

all resuIts were normalized to the abundance of cyclophilin mRNA. (n= 7 for ASMase
group. P < 0.05 compared to ASMase“+ control eye).

215

 

 

NSMase

3 * [:1 Control eye
2-5 I IIR eye

Transcript/ Cyclophilin

   

ASMase +5“ ASMase "’

Figure 5.2. Retinal gene expression of NSMase in the [IR retinopathy model. RT-
PCR analysis of NSMase in I/R injured retinas isolated from ASMase“+ and ASMase""
mice 2 days following retinal ischemia-reperfusion injury. The relative amounts of
mRNAs were calculated by using the comparative CT method. Cyclophilin was used as a
control and all results were normalized to the abundance of cyCIOphilin mRNA. (n= 7 for
ASMaseW group, n= 3 for ASMase‘/' group. *P < 0.05 compared to ASMase“+ control
eye, ASMase”+ I/R eye and ASMase"" control eye).

216

 

ICAM-l
C 0.15
E E] Uninjured eye
2 . .
g 0-1 Il/R Injured eye
0
E
'c 0.05
O
U)
C
('3 FL'
1- 0 _ ,- _ .
ASMase I” ASMase"'
B.
0.3 , VCAM-l
E
E
Q.
(_3
0
>5
0
b
.9-
b
(D
C
9
l—.

 

ASMase 1" ASMase "'

Figure 5.3. Retinal gene expression of cellular adhesion molecules in the HR
retinopathy model. RT-PCR analysis of ICAM-1(A) and VCAM-l (B) in UK injured
retinas (black bars) isolated from ASMase“+ and ASMase'/' mice 2 days following retinal
ischemia-reperfusion injury. The relative amounts of mRNAs were calculated by using
the comparative CT method. Cyclophilin was used as a control and all results were
normalized to the abundance of cyclophilin mRNA. (n= 7-9 for ASMase“+ group, n= 5
for ASMase'/' group. *P < 0.05 compared to ASMase“ control eye and ASMase'/' control
and I/R eye).

217

 

 

 

 

  

     

 

 

 

 

IL-lB TNFa
.E 06 0.02
E *
3 0.015 1
g 0.4 1 I
O :
E
'z 0.2
8 1'
c:
(0
I: 0 ——~---— -- - —*—- ----—~—— ---
ASMase +’+ ASMase"‘ ASMase +/+ ASMase 4’
C.
VEGF

.5 1

E

o.

2

o

>

0 0.5

:1

.9-

'5

(D

C

re

f: 0

ASMase 7“ ASMase "'

Figure 5.4. Retinal gene expression of inflammatory/angiogenic molecules in the I/R
retinopathy model. RT-PCR analysis of IL-lB (A), TNFa (B) and VEGF (C) in I/R
injured retinas isolated from ASMase“ and ASMase“ mice 2 days following retinal
ischemia-reperfusion injury. The relative amounts of mRNAs were calculated by using
the comparative CT method. Cyclophilin was used as a control and all results were
normalized to the abundance of cyclophilin mRNA. (n= 6-9 for ASMase“+ group, n= 5
for ASMase"" group. ‘P < 0.05 compared to ASMase” control eye and ASMase"' control
and I/R eye).

218

ASMase i”

 

I Uninjured eye I I l/R injured eye I

‘u-v weir, " ‘.‘ ."‘.1
. .

--‘ A». ‘i

’. ‘. 'l g\..._' 1‘ _>.'4
‘I ‘ ''r ‘l"- ‘ "t' “A '1“ ‘A.(:”‘ .‘ ‘ * Q .

sex;

 

Figure 5.5. Retinal capillary degeneration in retinal ischemia/reperfusion in [IR
retinopathy mouse model. Acellular capilla occurrence (black arrows) in the retina
isolated from (A) ASMase“+ and (B) ASMase" ' mice 7 days following retinal I/R injury.

219

ASMase 4'

 

Uninjured eye l/R injured eye

‘~.)\l’

I . f

       

Figure 5.5 (continued). Retinal capillary degeneration in retinal
ischemia/reperfusion in [IR retinopathy mouse model.

220

_s
01

.L
O

01

   

Acellular capillaries/mm2

 

 

0 “fin-“ __ __,
ASMase ”L” ASMase "'

 

[j Uninjured eye

I IlR injured eye

 

 

 

Figure 5.5 (continued). Retinal capillary degeneration in retinal
ischemia/reperfusion in [IR retinopathy mouse model. (C) Quantification of the
number of acellular capillaries fi'om five independent sets of mice, with a total of seven
mice per ASMase” group and six mice per ASMase'I' group. At least eight fields of
retina were counted in duplicates by two independent investigators; *P<0.05; red bar =
550nm, black bar = 110nm.

221

60
50
40

30 *'
20 -

 

10

Number of endothelial cell
nuclei (neovascularization)

 

 

 

 

- ASMase W

.__J

E! ASMase "' I

Figure 5.6. Retinal neovascularization in oxygen-induced retinopathy in mice.
Quantification of the endothelial cell nuclei representing abnormal retinal
neovascularization in the ASMase” and ASMase'l' mouse retina. (n= 8 per ASMase”
group and n= 6 per ASMase’/' group; 1IP<O.05 compared to number of endothelial cell
nuclei in ASMase“+ mouse).

222

5. Discussion

A large body of literature supports the role of ASMase activation in response to
various stress stimuli, including pro-inflammatory cytokine stimulation (TNFa and IL-
IB), oxidative damage (27), advanced glycation end products (AGES) (28) or protein
kinase C6 (PKCS) activation (29). These biochemical pathways are activated in the
diabetic retina and can damage the retinal vasculature, resulting in formation of diabetic

retinal microangiopathy lesions (30).

To corroborate the role of ASMase in retinal inflammation and microangiopathy
in vivo, we used a mouse model of retinal ischemia-reperfusion. Retinal ischemia-
reperfusion (I/R) is a well established model that closely resembles diabetic retinopathy
(8). We addressed the function of ASMase in retinal microangiopathy by making use of
an ASMase'l' mouse model of Niemann-Pick disease (NPD). As the ASMase"' mouse
model of NPD develops neurodegenerative complications early in life and has a short
lifespan, we are unable to maintain a diabetic ASMase'l' mouse model for a duration of
time sufficient to ensure the development of diabetic retinal complications. Therefore, in
the ASMase'/' mouse retinopathy was modeled by retinal I/R injury, which permits us to
examine the extent of retinal inflammation and capillary degeneration two days and seven

days after injury, respectively (8).

A possible mechanism for capillary degeneration seen in the retinal I/R model
could be activation and adhesion of leukocytes to the vasculature, inducing cell death

signaling in capillary endothelial cells. Indeed, our study shows that retinal I/R injury

223

induces significant upregulation of ICAM-1 and VCAM-1 expression in the wild type

retina (Figure 5.3 A,B).

Another possible mechanism for capillary degeneration is pro-inflammatory
cytokine secretion by various retinal tissues, such as glial cells or retinal pigment
epithelial cells, in the injured retina. We have found significant upregulation of IL-lfi,
TNFa and VEGF gene expression (Figure 5.4 A,B,C) in wild type retinas isolated from
I/R injured eyes. These inflammatory mediators can promote the activation and
degeneration of capillary endothelial cells with subsequent profound impact on vascular
permeability and viability. Indeed, in the present study, retinas with UK lesions exhibited
a significant number of acellular, degenerated capillaries seven days after initiation of
retinal injury; this time-point corresponds with late stage development of diabetic
retinopathy. More importantly, ASMase genetic deficiency decreased pro-
inflammatory/angiogenic mediator’s gene expression and thus prevented the development
of acellular capillaries in the I/R injured retinas. Therefore, ASMase emerges as a critical
initial responder in pro-inflammatory/angiogenic molecules-induced retinal

microvascular degeneration in a mouse model of retinal I/R injury.

Interestingly, we observed an increase in NSMase gene expression in the I/R
injured retinas from ASMase"' mice. Both ASMase and NSMase are considered key
mediators of stress-induce ceramide generation and cellular injury and death; yet, these
enzymes have individual requirements for their activation and regulation. ASMase
functions optimally at a pH of 4.5-5.5, whereas optimal NSMase activation occurs at pH
of 7.4 (31). NSMase upregulation in the ASMase'l' retinas with I/R lesions might try to

compensate the absence of ASMase and mediate the pathological processes that result in

224

 

cellular apoptosis. However, the acidic milieu generated in the retina by the ischemia-
reperfusion injury may preclude fully activation of NSMase and thus downplay the role
of NSMase in promoting the microvascular lesions in the UK model of retinopathy.
Importantly, the retinas isolated from ASMase'l' mice with retinal I/R lesions were
significantly protected against the inflammatory status and development of acellular
capillaries, suggesting that ASMase rather than NSMase could represent a key player in

mediating vascular damage in the I/R model of retinopathy.

In a recent study, Connor et al. demonstrated that docosahexaenoic acid (DHA)
supplementation by dietary or genetic means confer significant protection against
development of retinal pathological neovascularization in an OIR mouse model of
retinopathy. As shown in previous chapters, ASMase is increased in diabetic retina and
DHA returns ASMase to the basal levels in diabetic retina. Now we demonstrate that
ASMase'/' mouse retinas are also significantly protected against pathological

angiogenesis in the OIR model of retinopathy.

In conclusion, ASMase genetic deficiency renders protection against retinal
capillary degeneration and ocular neovascularization in accelerated retinopathy mouse

models.

225

6. References

10.

Hueber, A.O., Bernard, A.M., Herincs, Z., Couzinet, A., and He, HT. 2002. An
essential role for membrane rafis in the initiation of Fas/CD95-triggered cell death
in mouse thymocytes. EMBO Rep 3:190-196.

[(0, Y.G., Lee, J.S., Kang, Y.S., Ahn, J.H., and Sec, J.S. 1999. TNF-alpha-
mediated apoptosis is initiated in caveolae-like domains. J Immunol 162:7217-
7223.

Martin, M.U., and Wesche, H. 2002. Summary and comparison of the signaling
mechanisms of the Toll/interleukin-l receptor family. Biochim Biophys Acta
1592:265-280.

Horinouchi, K., Erlich, S., Perl, D.P., Ferlinz, K., Bisgaier, C.L., Sandhoff, K.,
Desnick, R.J., Stewart, C.L., and Schuchman, EH. 1995. Acid sphingomyelinase

deficient mice: a model of types A and B Niemann-Pick disease. Nat Genet
10:288-293.

Garcia-Barros, M., Paris, F., Cordon-Cardo, C., Lyden, D., Rafii, S., Haimovitz-
Friedman, A., Fuks, Z., and Kolesnick, R. 2003. Tumor response to radiotherapy
regulated by endothelial cell apoptosis. Science 300:1155-1159.

Smith, E.L., and Schuchman, EH. 2008. The unexpected role of acid
sphingomyelinase in cell death and the pathophysiology of common diseases.
Faseb J 22:3419-3431.

Wong, M.L., Xie, B., Beatini, N., Phu, P., Marathe, 8., Johns, A., Gold, P.W.,
Hirsch, E., Williams, K.J., Licinio, J., et al. 2000. Acute systemic inflammation
up-regulates secretory sphingomyelinase in vivo: a possible link between
inflammatory cytokines and atherogenesis. Proc Natl Acad Sci U S A 9728681-
8686.

Zheng, L., Gong, B., Hatala, D.A., and Kern, TS. 2007. Retinal ischemia and
reperfusion causes capillary degeneration: similarities to diabetes. Invest
Ophthalmol Vis Sci 48:361-367.

Agardh, C.D., Gustavsson, C., Hagert, P., Nilsson, M., and Agardh, E. 2006.
Expression of antioxidant enzymes in rat retinal ischemia followed by

reperfusion. Metabolism 55:892-898.

Hangai, M., Yoshimura, N., and Honda, Y. 1996. Increased cytokine gene
expression in rat retina following transient ischemia. Ophthalmic Res 28:248-254.

226

 

ll.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

J0, N., Wu, GS, and Rao, NA. 2003. Upregulation of chemokine expression in
the retinal vasculature in ischemia-reperfusion injury. Invest Ophthalmol Vis Sci
44:4054-4060.

Kuriyama, H., Waki, M., Nakagawa, M., and Tsuda, M. 2001. Involvement of
oxygen free radicals in experimental retinal ischemia and the selective
vulnerability of retinal damage. Ophthalmic Res 33:196-202.

Vorwerk, C.K., Lipton, S.A., Zurakowski, D., Hyman, B.T., Sabel, B.A., and
Dreyer, EB. 1996. Chronic low-dose glutamate is toxic to retinal ganglion cells.
Toxicity blocked by memantine. Invest Ophthalmol Vis Sci 37:1618-1624.

Yoneda, S., Tanihara, H., Kido, N., Honda, Y., Goto, W., Hara, H., and
Miyawaki, N. 2001. Interleukin-lbeta mediates ischemic injury in the rat retina.
Exp Eye Res 73:661-667.

Gottlieb, R.A., and Engler, R.L. 1999. Apoptosis in myocardial ischemia-
reperfusion. Ann N Y Acad Sci 874:412-426.

Levade, T., Auge, N., Veldman, R.J., Cuvillier, 0., Negre-Salvayre, A., and
Salvayre, R. 2001. Sphingolipid mediators in cardiovascular cell biology and
pathology. Circ Res 89:957-968.

Bonne, C., Muller, A., and Villain, M. 1998. Free radicals in retinal ischemia.
Gen Pharmacol 30:275-280.

Osborne, N.N., Casson, R.J., Wood, J .P., Chidlow, G., Graham, M., and Melena,
J. 2004. Retinal ischemia: mechanisms of damage and potential therapeutic
strategies. Prog Retin Eye Res 23:91-147.

Das, A., and McGuire, PG. 2003. Retinal and choroidal angiogenesis:
pathophysiology and strategies for inhibition. Prog Retin Eye Res 22:721-748.

Dorrell, M., Uusitalo-Jarvinen, H., Aguilar, E., and Friedlander, M. 2007. Ocular
neovascularization: basic mechanisms and therapeutic advances. Surv Ophthalmol
52 Suppl 1:83-19.

Smith, L.E., Wesolowski, E., McLellan, A., Kostyk, S.K., D'Amato, R., Sullivan,
R., and D'Amore, RA. 1994. Oxygen-induced retinopathy in the mouse. Invest
Ophthalmol Vis Sci 35:101-111.

Marchesini, N., and Hannun, Y.A. 2004. Acid and neutral sphingomyelinases:
roles and mechanisms of regulation. Biochem Cell Biol 82:27-44.

Pavoine, C., and Pecker, F. 2009. Sphingomyelinases: their regulation and roles in
cardiovascular pathophysiology. Cardiovasc Res 82:175-183.

227

 

24.

25.

26.

27.

28.

29.

30.

31.

Caballero, S., Sengupta, N., Afzal, A., Chang, K.H., Li Calzi, S., Guberski, D.L.,
Kern, TS, and Grant, MB. 2007. Ischemic vascular damage can be repaired by
healthy, but not diabetic, endothelial progenitor cells. Diabetes 56:960-967.

Laver, N.M., Robison, W.G., Jr., and Pfeffer, BA. 1993. Novel procedures for
isolating intact retinal vascular beds from diabetic humans and animal models.
Invest Ophthalmol Vis Sci 34:2097-2104.

Agardh, E., Gustavsson, C., Hagert, P., Nilsson, M., and Agardh, CD. 2006.
Modifying a standard method allows simultaneous extraction of RNA and protein,
enabling detection of enzymes in the rat retina with low expressions and protein
levels. Metabolism 55:168-174.

Dumitru, C.A., Zhang, Y., Li, X., and Gulbins, E. 2007. Ceramide: a novel player
in reactive oxygen species-induced signaling? Antioxid Redox Signal 9:1535-
1540.

Patschan, S., Chen, J ., Polotskaia, A., Mendelev, N., Cheng, J ., Patschan, D., and
Goligorsky, MS. 2008. Lipid mediators of autophagy in stress-induced premature
senescence of endothelial cells. Am J Physiol Heart Circ Physiol 294:H1119-
1129.

Zeidan, Y.H., and Hannun, Y.A. 2007. Activation of acid sphingomyelinase by
protein kinase Cdelta-mediated phosphorylation. J Biol Chem 282:11549-11561.

Fong, D.S., Aiello, L.P., Ferris, F.L., 3rd, and Klein, R. 2004. Diabetic
retinopathy. Diabetes Care 27:2540-2553.

Kolesnick, R. 2002. The therapeutic potential of modulating the
ceramide/sphingomyelin pathway. J Clin Invest 110:3-8.

228

 

Conclusion

This study demonstrates the role of sphingomyelinases, key regulatory enzymes
of sphingolipid metabolism, in modulating the inflammatory process in human retina]

endothelial cells, the target vasculature affected by diabetic retinopathy.

We showed that proinflammatory cytokines activate sphingomyelinases and
sphingomyelinase downregulation (by pharmacologic inhibitors, DHA or gene silencing)
prevents cytokines-induced adhesion molecules expression in human retinal endothelial
cells. In suport of sphingomyelinases downregulation and more importantly,
downregulation of caveolar ASMase by DHA, we found a decreased
ceramide/sphingomyelin ratio in caveolae lipid domains; moreover, DHA pretreatment of
HREC prevented cytokine-induced increase in ceramide/sphingomyelin ratio in these

specialized lipid domains.

Using in vivo studies we confirmed that ASMase, but not NSMase expression was
significantly upregulated in diabetic rat retinas and closely correlated with the
inflammatory status and vessel loss in diabetic retinal tissue. Importantly, ASMase

increased specifically in the retinal vasculature from I/R injured eyes.

DHA supplementation significantly prevented increased ASMase expression,
inflammatory status and vessel loss in the diabetic retina and prevented ASMase
upregulation in the retinal blood vessels. ASMase genetic deficiency prevented
upregulation of inflammatory/angiogenic molecules, vessel loss and pathological
neovascularization in mouse retinas with accelerated models of retinopathy, and thus

recapitulated the protective effects of DHA on diabetic retinas.

229

 

Therefore, our data suggests that ASMase, the key regulatory enzyme of
sphingolipid metabolism, is a novel mediator and a promising therapeutic target for the
prevention of retinal vascular inflammation and diabetic microvascular complications.
Inhibition of ASMase by genetic means, pharmacologic inhibitors or DHA
supplementation could be beneficial in preventing the development of microangiopathy
lesions in diabetic retinopathy. This finding is particularly attractive since currently
available approaches for the treatment of diabetic retinopathy do not fully prevent or

reverse ocular pathology.

However, further studies are needed to elucidate and confirm the role of ASMase
in the development of diabetic retinal inflammation and microangiopathy. One of the
studies would be conducted in type 1 and type 2 diabetic rats, and ASMase inhibition will
be achieved at the retinal level with inhibitors (i.e. desipramine, imipramine) that can be
administered by intravitreal injection or slow-release lipid-based inhibitors implanted
preretinal. To directly test the role of vascular ASMase in mediating retinal vessel loss,
another study can be performed using vascular ASMase deficient mice that will be made
diabetic by streptozotocin injection and retinal vessel damage would be assessed after
more then 6 months after induction of diabetes. All these studies may provide new clues
and let us understand more comprehensively about the role of ASMase, and particularly

its vascular localization, in the development of retinal vascular damage in diabetes.

230