0

Abnormal

Free Fatty

Concentrations
Nl%l?NA

lJ.224 INSERM

et Laboratoire

France and tService

Acids and Cortisol

in the Serum

CHRISTEFF,*
CHRISTOPHE
MATHERON,t
PIERRE

of AIDS

Patients

MICHON,?
GUY GOERTZ,* JACQUELINE
HASSID,* SOPHIE
MARIE GIRARD,t J.P. COULAUDt
and EMMANUEL
A. NUNEZ*z

de Biochimie

de Pathologie

0277-.i379/RR$3.u0+0.00
1988 PtTqamn Press plr

B, Faculte et Hipita

infectieuse

et tropicale,

Xavier Bichat,

Hipita

16 et 46 rue Henri Huchard,

Claude Bernard,

75018

10 au de la Porte &Aubervilliers,
Paris,

serum free fat&

Abstract-The

steroid and 17-oxosteroid

ning and AIDS-II:

acid (FFA),

concentrations

end phase)

cortisol and urinary creatinine,

complex

(ARC)

patients

France

17+ydzoxycortico-

of acquired immunedejiciency syndrome (AIDS-I:

and AIDS-related

Paris,
75019

begin-

were determined.

Both groups were compared

to a control group (healthy men).

ARC and AIDS-I patients. The ratios of stearic (C18:O) to oleic (C18:l) acid were
75%, P < 0.01 (ARC) and 45%, P < 0.05 (AIDS-I) greater than normal, due to a decrease
in the relative percentage of monounsaturatedfatty
acids by 25%, P < 0.001 (ARC) and 20%,
P < 0.01 (AIDS-I). In contrast, the relativepercentage ofpolyunsaturatedfatp
acids was 85%
greater than normal (P < 0.001) in ARC and 100% greater than normal (P < 0.001) in
AIDS-I

patients.

Total FFA levels did not diff

from controls. Serum cortisol levels were 35%

(P < 0.01) above normal in ARC and 60% (P < 0.001) above normal in AIDS-I patients.
Urinary 1I-hydroxycorticosteroids
and 17-oxosteroids were very low (2-3-fold lower than normal
values, P <O.OOl) in both groups of patients. Urinary creatinine did not differ from controls.
In AIDS-II patients the total FFA concentration was below normal 35% (P < 0.01) and
the stearic/oleic acid ratio was 50% above normal (P < 0.05). The relative percentages of
monounsaturated and polyunsaturated fat& acids in this group were similar to those of controls.
Serum cortisol concentrations were signiJcantly higher, 50% (P < O.OOl), but he urinary 17hydroxycorticosteroids and 17-oxosteroids were 2-fold lower (P < 0.001) than those of controls.
Urinary creatinine did not diff
These sigr@cant

from

diJrferences from

controls.

normal may be implicated

and could represent not only a good index of diagnosis

therapeutic approach

in the pathophysiolou

and prognosis,

to the disease.

INTRODUCTION
RECENT studies

have

acid

conrcnts

cells

arc abnormal

syndrome
tent
higher
free

than

clear

adult
when

between

cells

those

results

are

and
[l]).

oleic

acid

The

in AIDS

sufferers

of various

lymphocyte

blood

viability

blood

ratio

patients

can cause
immune
and

of the UFA
which

when

enhances

(UFA)
added

the serum

as the

particular
measure

ratio

concentrations

those

is significantly

various

may,

This

study

of each
stages

levels of FFA,

class

and

17-hydroxycorticosteroids

but not in anti-

cortisol

of the disease.
and

acid.

as is that
with linoleic

effect on lympho-

and cortisol

patients

significance.
the serum

together

its negative

or symptom-

of AIDS

such as linolcic

is dose-dependent,

virus

(HIV)

in vitro by free

is mimicked

The action
acid, greatly

is

action

fatty acids

of cortisol

is particularly

contents

therefore,

of

be of

was designed

to

both the total and

from

Urinary

patients

at

creatininc,

17-oxosteroids

were

also determined.
extracts

of plasma

considerable
responses,

the

cells

[2]. This

unsaturated

cyte viability [3].

The free fatty acid (FFA)

expressed

HIV positive patients.

It is also known that ethanolic
fication

white

subjects

men. This abnormality

the

the fatty

immunodcficiency
from

of normal

index

AIDS

and

immunodeficiency

lower than normal

from

that

[ 11. The oleic acid con-

extracted

the stearic

(saturation

blood

in acquired

lipids

ratio

shown

patients

anti-human

positive

clearly

of red

(AIDS)

of the

of AIDS

but also indicate new

MATERIALS AND METHODS

modiespecially

Patients

T-CD4+/T-CD8+

Sixty HIV seropositive

in-patients
or out-patients
Hospital,
Paris, between
1987: 16 patients
with

Accepted 25 February 1988.

iAuthor to whom correspondence should be addressed.
I179

patients
were studied as
at the Claude
Bernard
March
and
AIDS-related

September
complex

1180

N. Christeff

(ARC),

25 patients

with

early

since less than

3 months-AIDS-I)

with

AIDS

advanced

(diagnosis

AIDS

(diagnosis

and

19 patients

since

more

Serum free fat& acid (FFA)

than

or the diagnosis

(KS) in four and both 01 and KS in 11.

adult

male,

50 European

from Africa

one risk factor

drug

addicts,

Kaposis

age 35 years

patients

(9 patients)

least

in 29 patients,
mean

and

or Brazil

(49 homo-

had at

or bisexual

was

carefully

men,

3 heterosexuals).

estimated

for

each

group

of

ARCpatients:
loss >lO%

no loss (45%),

loss <lo%

was

removed

extracts
up

in

LH20

0.5 ml

with

ated

to

loss >lO%

loss <lo%

loss >lO%

noloss

(15%),

All patients

receiving

were excluded.

cotrimoxazole,
conazol)

azydothymidine

Anti-infection

sulfadiazine,

In contrast

before

samples

were taken.

the blood

patients

drugs.
Anxiolytic
drugs,
particularly
diazepines,
were very often administered

patients

from

were

both

groups

compared

group

to those

extremes

obtained

25-50

years)

Technologies).

The serum
were

by 4C/min
240C

were

from

Fused
The

to 230C. The

and

internal

the detector

standard

ratios.

determined

values

Concentrations

in a Packard

cortisol

determined

and urinary
by

free cortisol

fluorescence

604

Division

levels

polarization

an immunofluorimetric
immunoassay,
using TDX System cortisol from Abbott

seropositive

adult males

acids

(United

Beker,

(WCOT

175C for the first 5 min

to standard

of fatty

on a Packard
(Packard,

Cortisol determination

received
the benzoto all three

of HIV

(n = 38) ofhealthy

age 40 years,

keto-

all AIDS-I

compared

boron

methylated

25 m x 0.32 mm).

was

were

as

were
evapor-

in

The

column

thereafter

temperature

(such

for almost

very few of the ARC

groups.
The data

was

or inter-

drugs

pyrimethionine,

had been prescribed

or -11 patients

temperature

FFA
were

methylated

419

CB,

temperature
was 260C.
Peak area ratios and

(75%).

solvent

on Sephadex

chromatographed

CP-WAX-52

injector

loss < 10% (lo%),

placed

(Merck).

The

to dryness,

Extracts

a capillary

and increasing

(30%).

AIDS-Zlpatients:

and

model

using

aqueous

(-20C).

evaporated

and

were

solvent

the

(0.5 X 6 cm).

dryness

acids

column

(50%),

freezing

3 ml of solvent.

trifluoride-methanol
fatty

and

of chromatography

95:5)

microcolumns

eluted

Silica
no loss (20%),

by

samples

(0.5 ml) was

5 ml organic
v/v)

were pooled,

(benzene-ethanol,

to serum

sample

with

phase

U.S.A.)

(50%),

(5%).

AIDS-Zpatients:

control

times

Chromatograph,

patients.

feron

three

Each

acetate/cyclohexane

taken

10 immigrants

These patients
had received
no parenteral
or
enteral artificial nutrition,
The percentage
ofweight
loss

acid was added

standard.

(ethyl
organic

(range

(1 patient)

4 blood-recipients,

extracted

was oppor-

sarcoma
23-57),

(01)

of AIDS

infections

All were

Heptadecanoic
as an internal

tunistic

extraction and gas chromato-

graphic analysis

months-AIDS-II).
The criteria

et al.

method
Diagnostics

(U.S.A.).

(mean

receiving

no

Determination of urinary 17-hydroxycorticosteroids and 17oxosteroids

medication.
It was not possible
the patients

to perform

and controls,

all the tests on all

their number

for each type of determination

is indicated

Urinary
steroids
of Silber

in the tables.

17-hydroxycorticosteroids

were estimated
and Porter

and

respectively

17-0x0-

by the methods

[5] and Zimmermann

[S].

Statistical analysis
Blood and urine samples
All blood
allowed

samples

to coagulate

centrifugation
Serum

The
were
before

obtained
separation

(3000 rpm/lO

samples

and

of serum

by

min at 4C).

were treated

by creatinine

[4], was collected.

Fat0 acids standards

Myristic, palmitic, stearic, heptadecanoic,
palmitoleic (n-7), oleic (n-9), linoleic (n-6), linolenic
(n3), arachidonic
docosahexaenoic
Sigma Chemical

were
were:

(n-6), docosatetraenoic
(n-6) and
(n-3) acids were purchased
from
Company.

are

reported

as

means

t test was used to compare

considered
*P < 0.05,

significant

when

**P < 0.01,

S.E.M.

mean.

Results

the probabilities

***P

< 0.001.

RESULTS

30 min at 56C and

stored at -20C until assayed.

Twenty-four
hours urine, checked
determination

at 8 a.m.

data

Students

Serum concentration of free fat4

acids (FFA)

in different

groups of ARC and AIDS patients

The serum levels of total FFA, the relative percentages
of saturated,
mono- or polyunsaturated
fatty acids

and the ratios

of stearic

(C18:O)

acid (C 18: 1) in each group of ARC

patients are shown in Table 1.

and

to oleic
AIDS

The serum levels of total FFA were significantly

lower than normal
(35%, P < 0.01) only in the
AIDS-II
patients.
Total
FFA in the ARC and
AIDS-I were not different from normal controls.

Free Fat0

Acids and Cortisol in AIDS

1181

Table 1. Serum FFA levels, relative percentage of the various classes of FFA and ratio of stearic (C 18:0) ta oleic (C 18: 1) acids in ARC and AIDS
patients
FFA(%)
C:18:0

FEA
(mM/l)

Saturated

Monounsaturated

Diunsaturated

Polyunsaturated

C:18:1

Controls
n = 21

0.87 2 0.1

30 2 3

38 t

1.4

26.2 + 1.5

5.8 2 0.5

0.2 k II.02

ARC
n=

13

0.75 + 0.1

34 2 1.2

30 t

1.4***

25 2 1.3

10.8 + 0.9***

0.35 + 0.02**

AIDS-I
n= 11

0.85 ? 0.1

34.6 -r- 1.6

31.2 ? 1.7**

22.5 r 1.1

11.7 + 1***

0.29 + 0.01*

AIDS-II
rl= 12

0.58 + 0.08**

34 t 1.6

34.6 2 2.1 (N.S.)

24.2 ? 1.3

7.2 + 0.7 (N.S.)

0.30 2 0.05*

Gas chromatographic

quantification

Means ofdeterminations
P < 0.05, **p < 0.01,

The relative
(C18:l

and

percentages

of monounsaturated

C16: 1) were

(P < 0.001)
normal

of FFA serum levels

k S.E.M.
***p < 0.001.

in ARC

25%

lower

patients

and

(P < 0.01) in AIDS-I

ence was particularly

By contrast,

the relative

rated

acids

fatty

20%

normal

lower

patients.

marked

FA

than

The creatinine

percentages

and

differ-

DISCUSSION
The

C22:6)

in

data

trol values,
is decreased

(P < 0.001)

AIDS-I

(PUFA)

patients.

The

relative

percentages

diunsaturated

FA were

from

in any

controls

stcaric

not

in

of saturated

significantly

of the groups.

P < 0.01) and AIDS-I

(75%,

P < 0.05) than

(50%,
There

higher

The

in

ticularly

of the various

in the rela-

classes

of FFA in the

patients.

the

patients.

cells and the surrounding

the result

of a transfer

difficult,
tration

hydroxycorticosteroid
controls,
ARC
shown in Table
The
cantly
P<

serum
higher

O.Ol),

and
2.
cortisol

and

17-oxosteroid

AIDS-I

(60%,

values
P<

levels

-11 patients

concentrations

than control
AIDS-I

and

were

in the ARC
0.001)

and

of
are

signifi(35%,
AIDS-

II (50%,
<O.OOl)
patients.
In contrast,
the
urinary levels of 17-hydroxycorticosteroids
and 17oxosteroids
were very significantly
lower than normal
AIDS

(2-3-fold,
patients.

P < 0.001)

in the ARC

and

all the

can be

these

at present,
of PUFAs

patients.

inhibition
acids

to explain

the

why

cells.
the

in the

concen-

of the various

may provide
different

in

of highly

patients,

which

significantly

lower

AIDS-II
than

normal

of

PUFA
intra-

the data

to interpret
this result.
The situation
is quite
critical

of

It is

serum

of the cellular

and activities
lipases

may be

of the rcleasc

from

is enhanced

Determination

concentration
extracellular

serum

of olcic acid from the serum

hydroxycorticosteroids and 17-oxosteroids in different groups

17-

par-

results

in a recent study
to be significantly

the blood

of ARC and AIDSpatients

creatinine

abnor-

for C18:l.

decreased
in the red and white blood cells of AIDS
patients. The existence ofsuch a reciprocity
between

fatty

urinary

acid

increased,

These

with those obtained

this ratio was found

strong

and

These

particularly

C18:O/C18: 1 ratio is also clearly

to the cells and/or

cortisol

fatty

patients.

significant,

in the ARC

the con-

FA content

in the AIDS-related

and AIDS-I

monounsaturated

serum

the total

from

polyunsaturated

Serum concentrations of cortisol and urinary creatinine, 17-

The

although

monounsaturated

is increased

(ARC)

compared
[l] where

and -11 patients

differences

and

are highly

of

in controls.

were no significant

tive percentages
AIDS-II

ratio

that,

are not different

content

complex
malities

different

The

acid to oleic acid was significantly

ARC

and

show

the strum

those of
patients

(P < 0.001)

clearly

FFA concentrations

these same two groups were almost double

controls
(5.8%),
being
10.8% in ARC
11.7%

patients

controls.

than

This

of polyunsatu-

C22:4

and AIDS

from normal

for oleic acid ( C18: I).

(C20:4,

and

levels in ARC

were not different

or

required

the

serum
have

concentration

a
of

total free fatty acids. This may result from major

metabolic
perturbations,
with an increase in the use
of lipids for energy, as is usually observed in starved
and severely stressed patients in the terminal phase
of

the

disease

The

other

lipid

perturbations

1182

N. Christeff et al.
Table 2. Serum cortisol and urinary 17-hydroxycorticosteroid,

17-oxosteroid

and creatinine concentrations in ARC

and AIDS patients

17-Hydroxycorticosteroids
(pmo1/24 hr)

Cortisol
(nWI)

11.62 0.6

17-Oxosteroids
(pmol/24 hr)

Creatinine
(mmoU24 hr)

44.6 2 2
n= 18

13 + 0.7
lz= 18

341 k 28
?I= 17

n=

ARC

471 2 34**
n= 16

6.7 2 l***
t, = 6

15.6 k 6.7***
n=6

13 ? 2.7
n=6

AIDS-I

537.5 2 32***
n = 25

5.5 + 3.6***
n = 13

21.5 2 4***
Il= 13

10.7 t 1.2
?l= 13

AIDS-II

516.6 -c 44***
n= 19

5.5 k 1.4***
n=7

20.6 ? 7.7***
n=7

12.8 2 1.4
n=7

Controls

18

Serum cortisol levels were determined

by immunofluorimetric
method using TDX systrm cortisol (Abbott
Diagnostics).
Urinary creatinine levels were determined by the Jaffi reaction, urinary 17-hydroxysteroids
and 17-oxosteroids
respectively by the methods of Silber and Porter and Zimmermann.
Values a& mkans 2 S.E.M.; **P < 0.01; ***P < 0.001

observed

in the ARC

this AIDS-II
ratio which

except

is still slightly

The combination
cortisol

and AIDS-I

group,

and

arc not found

urinary

of serum

concentration

of

17-

hydroxycorticosteroids
in ARC and AIDS-I
and
-11 patients, without any modification
of the urinary
creatinine,
can be compared
with the in vitro results
showing
AIDS

that ethanol
patients,

the serum

extracts

cancer

ofsera

and

of newborns

from ARC and

cirrhosis

all inhibit

More
hormonal

investigations
are needed
status of these patients:

-Dynamic

increased.

of high concentrations

low

in

that of the C 18:0/C 18: 1

patients

and

the metabolism

of

stimulation

various

steroids

by ACTH

produced

to define

the

exploring

the

by adrenals

mineralocorticoids,
androgens)
-Determination
of the concentration
steroid

bindingglobulin

catabolism.
-Determination

which

of the urinary

(gluco-,

of the cortico-

control

the cortisol

free cortisol.

Preliminary
results of the urinary
free cortisol
obtained
in AIDS patients
(264 k 70 nM, n = 7)

glucocorticoids
by lymphocytes
[7,8].
Indirect evidence suggests that one of the ethanol

show that this compound

is significantly
enhanced
by comparison
with
controls
(98.3 + 12.5 nM,

soluble

compounds

n = 9). Furthermore,

cortisol

metabolism

responsible

acid [3]. The existence

patients

in the urine

can be explained
steroid

by impaired

viability,

and

acid

linoleic

although

when

alone,

enced when these two compounds

The present
results
provide
pathophysiological

circumstances

in vitro may

correspond

during

between

(adrenal

information.

that

by cortisol

is more

influ-

are associated
cvidencc
that
produced

[3].
the

experin vivo

to a real

More

in vitro and

in vivo

as arachidonic,

docosatetraenoic

and

docosahexaenoic
acids may be involved
in this
phenomenon
in viuo instead
of linoleic acid, as their
concentrations
in the
patients
are high.

scra

of ARC

and

AIDS-I

work

stages

an

relevant

acid, linoleic acid [3], but the concentration

of this
UFA is unchanged
in ARC and AIDS patients.
such

various
that

is also

needed

observed
in AIDS
high concentrations

sera ofARC
as indicators

and hence

state observed

In addition
these results,
concentrations

infection
androgenic

to explain

how

(hypersecretion
with

adrenal

clinical

this
and/
and

insufficiency

[9], can act in association

with
of PUFAs and/or some of their

amplyifying

activity

suppressed

of

HIV

could also provide

contrasting

symptoms

the uri-

decreased

of the

of cortisol

or hypocatabolism),

metabolites,

of the

exploration

and testicular)

high concentration
biological

that

arc also significantly

function

studies is the fact that the immunosuppressive

effect
of cortisol
in vitro is potentiated
by the free fatty

PUFAs

the

suggested

pressive

One of the differences

the observation

17-oxosteroids

ofUFAs.

affected

used

nary

metabolism

it is clear from the same in vitro studies

Thus,

on

of AIDS

due to the high concentration

lymphocyte

imentally
situation.

effect

an UFA like linolcic

of a very low concentration

of 17-hydroxycorticosteroids
ofthe

for this

is probably

each others

immunosup-

aggravate

the immuno-

in AIDS.

to the pathophysiological
interest
of
knowledge
of the variation
in the
of glucocorticoids
and lipids in the
and AIDS patients could also be useful
of the stage of the disease, providing
a

good index for diagnosis

and prognosis.
The implication
of fatty acids and cortisol in the
pathophysiology
of AIDS raises the possibility
of a
new therapeutical
approach
to the disease,
which

Free Fatty Acids and Cortisol in AIDS

may involve induced modifications

of the lipidic
structure of the immune cells and a reduction in the
deleterious
effects induced on these cells by the
simultaneous

presence

glucocorticoids

of high

and PUFAs.

concentrations

of

Acknowledgements-This
work was supported by grants from
INSERM-AIDS
program. Ligue Nationale Franqaisr contre le
Cancer, Fondation pour la Recherche MCdicale and UER Xavier
Bichat (University Paris VII). The authors are grateful to 11rC.
Owen Parkes for helping in preparing
the manuscript
and to
Mrs P. Neant for secrrtarial assistance.

REFERENCES

5.
6.
7.

8.

9.

EJC

24:7-G

1183

Apostolov K, Barker W, Wood CB, Habib NA, Jeffries D, Forster SM. Fatty acid saturation
index in peripheral blood cell membranes ofAIDS patients. Lancet 1987, 21, 695.
Klein A, Malkin A, Pinkerton PH, Robinson JB, Rachlis A. Destruction of lymphocytes by
a non-viral fraction of HTLV III positive patients plasma. Turn Biol 1986, 7, 147-150.
Klein A, Bruser B, Malkin A. The effect of linoleic acid on the sensitivity of human
lymphocytes to cortisol and their capacity to catabolize the steroid. Biochem Cell Biol 1987,
65, 409413.
Bonsnes RW, Taussky HH. On the calorimetric determination
of creatinine by the Jaffe
reaction. J Biol Chem 1945, 158, 581.
Jayle MF. Dosage des 17 hydroxycorticost&oides
libres et conjugts urinaires par Silber et
Porter. Analyse des steroides hormonaux. In: Methodeset Dosages. Masson, Vol. 11, 143-147.
Jayle MF. Analyse des 17 &ostiroides.
Analyse des stiroides hormonaux. In: M&hodes et
Dosages. Masson, Vol. II, 1-17.
Klein A, Archer SM, Malkin A. Sera of patients with cancer or cirrhosis and of newborns
exhibiting inhibitory effects on the metabolism of cortisol by lymphocytes. JNCZ 1986, 77,
29-3 1.
Klein A, Bruser B, Robinson JB, Pinkerton PH, Malkin A. Effects of a non-viral fraction of
acquired immunodeficiency
syndrome plasma on the vulnerability oflymphocytes
to cortisol.
J Endocrinol 1987, 112, 259-264.
Greene LW, Cole W, Greene JB et al. Adrenal insufficiency as a complication of the acquired
immunodeficiency
syndrome. Ann Intern Med 1984, 101, 497-498.