- Email: [email protected]
Antibodies to lipoprotein lipase application to perfused heart
213
Biochirnica ct Bioph.vsica @ Elsevier/North-Holland
Acla,
489 (1977)
Biomedical
211-22‘1
Press
BBA 570X.1
ANTIBODIES
TO LIPOPROTEIN
APPLICATION
TO PERFUSED
LIPASE HEART
MICHAEL ‘2. SCHOTZ, JER-SHUNG TWU, MARY E. PEDERSEN, .4RLENE S. GARFINKEL and JAYME BORENSZTAJN a Research L’etcrans Administration, Wadsworlh Horpital Department of Medicine, UCLA School of Medicine and Nutritional Sciences, School of Public Health, University 90024, and a Department of Pathology, Pritzkcr School Chicago, Ill. 60637 (IJ.S.A.)
(Received
CHI-HONG
CHEN,
Center, Los Angeles, Calif 90073, the Division of Environmental and of California, Los Angeles, Calif. of Medicine, University of Chicago,
May 31st, 1977)
Summary An antibody was prepared against purified rat heart lipoprotein lipase. 1. This antibody showed marked species specificity, It inhibited almost totally the lipoprotein lipase activity from all rat tissues examined (i.e., heart, adipose, postheparin plasma, and mammary gland), while having no effect on the activity of lipoprotein lipase partially purified from rabbit, guinea pig and bovine heart and from bovine milk. The antibody also had no effect on the hepatic lipase activity of rat postheparin plasma. 2. After antibody to rat heart lipoprotein lipase was recirculated for 5 min through isolated rat hearts, little or no lipoprotein lipase activity could be detected in the perfusate during O-20 s of a subsequent non-recirculating perfusion with buffer containing 1 unit heparin/ml. 3. Following recirculation of antibody to lipoprotein lipase for 10 min and a non-recirculating perfusion with buffer for 2 min, the hearts no longer oxidized any significant amounts of 14C-labelled palmitate chylomicron triacylglycerol fatty acid to 14C0, during a 15-min perfusion. The data give compelling evidence that the functional fraction of lipoprotein lipase in hearts is at the endothelial cell surface accessible to lipoprotein lipase antibody.
Introduction Lipoprotein lipase is the enzyme required for the uptake of circulating protein triglyceride fatty acid by various extrahepatic tissues [I]. This established concept is based on several major pieces of evidence. First,
lipowelllipo-
215
protein lipase activity in each tissue correlates directly with the uptake of triglyceride fatty acid by that tissue [ 21. Further, release of a fraction of lipoprotein lipase activity from a tissue prevents the uptake [3]. Finally, adminstration of antisera to lipoprotein lipase prevents the removal of lipoprotein triglyceride from the plasma compartment [ 41. Lipoprotein lipase activities from different tissues (heart and adipose, for example) share the common properties of inhibition by salt and activation by a specific serum apolipoprotein. However, the fact that these enzyme activities appear to be regulated by different mechanisms raises the question as to whether their protein structures are similar. Recently lipoprotein lipase has been highly purified from rat heart tissue [ 5,6]. Employing antibody produced against this purified heart enzyme, the present study indicates that lipoprotein lipases from various tissues of the rat are indeed immunologically similar. The effect of lipoprotein lipase antibody on the physiological function of the enzyme, that is, the hydrolysis of circulating lipoprotein triglyceride by the intact heart, has also been studied. Materials and Methods Enzyme sources Sprague-Dawley rats (150-250 g) maintained on Purina Laboratory Chow were used. The tissue sources were hearts of male rats fasted overnight, epididymal adipose tissue from male rats fed ad libitum, and mammary gland of female rats lactating for 13 days fed ad libitum. Acetone powders were prepared from these tissues [7] and extracted with 5 mM barbital buffer, pH 7.2 (designated barbital buffer) [ 51. Postheparin plasma was obtained by intravenous injection of heparin (Calbiochem), 20 units/100 g body weight, into male rats which had been fasted overnight and were anesthetized with diethyl ether. Two minutes after injection, the rats were decapitated, the blood collected and the plasma obtained stored at -20” C. Enzyme purification Lipoprotein lipase was purified from rat heart acetone powders as previously described [5], except that elution of the enzyme from the heparin-Sepharose column was accomplished in the presence of 20% glycerol (v/v). Glycerol has been found to stabilize the enzyme activity. This partially purified preparation was employed in most of the experiments described. For preparation of antiserum, the lipoprotein lipase obtained by heparin-Sepharose chromatography was further purified by affinity chromatography on concanavalin A-Sepharose (Pharmacia) columns (1 X 5 cm) following the procedure of Bensadoun et al. [S]. The enzyme was eluted with barbital buffer containing 1.2 M NaCl, 20% glycerol (v/v) and 0.6 M oc-methyl-D-mannoside. With this procedure 50-60% of the enzyme activity applied to the column was recovered. After dialysis at 4°C for 24 h against a large excess of barbital buffer containing 0.05 M NaCl, the enzyme was concentrated by vacuum dialysis. It was stored at -20°C and used within 10 days. This purified lipoprotein lipase fraction showed three protein-staining bands on 1% sodium dodecyl sulfate polyacrylamide gel electrophoresis, one with a mobility corresponding to an apparent molecular
216
weight of 68 000; the other two were identical to bands found in the eluate when no protein was applied to the concanavalin A column. These latter bands apparently resulted from leakage of the matrix-bound ligand, concanavalin A, during chromatography [S] . The hepatic and extrahepatic lipases from postheparin plasma were separated by heparin-Sepharose [5]. The plasma was diluted several-fold with 0.15 M NaCl. After centrifugation at 3000 X g for 30 min at 4°C the supernatant fraction was adjusted to 0.4 M NaCl and a 5-ml sample applied to a column of heparin-Sepharose (1.5 X 15 cm) equilibrated with barbital buffer containing 0.4 M NaCl. The column was washed with the same buffer. The hepatic lipase activity was eluted with barbital buffer containing 0.75 M NaCl and 20% glycerol (v/v) and the extrahepatic lipase activity with the same buffer containing 1.5 M NaCl. Recovery of the total triglyceride hydrolase activity was go-100%. Antibody production Approx. 15-20 pg of the purified heart enzyme, emulsified with an equal volume of Freund’s complete adjuvant, was injected subcutaneously at multiple sites into the suprascapular region of a 60-kg female goat. This procedure was repeated twice at 2-weekintervals, substituting incomplete for complete Freund’s adjuvant. Ten days after the last injection, the goat was bled from the jugular vein and the serum separated. The IgG fraction was then isolated by ammonium sulfate precipitation and DEAE-cellulose chromatography [9]. Upon 1% sodium dodecyl sulfate polyacrylamide gel electrophoresis, the IgG fraction thus obtained displayed two protein-staining bands, corresponding to the heavy and light chains of immunoglobulin. The IgG fraction (6.4 mg/ml) was stored at -20°C in 10 mM phosphate buffer (pH 7.4) containing 0.015 M NaCl and 0.02% NaN, (w/v). The IgG fraction of serum obtained from a non-immunized goat was isolated and stored in the same manner as described above. Anticoncanavalin A activity of the IgG fraction was removed by passing the IgG fraction through a concanavalin A-Sepharose column, the unbound fraction constituting the purified antibody to lipoprotein lipase. Lipase assays and incubation conditions Lipoprotein lipase activity was assayed using as substrate serum activated [9,10-3H]oleic acid-labelled trioleoylglycerol emulsified with lysophosphatidylcholine [lo] or phosphatidylcholine [ll]. The assay for hepatic lipase activity was the same as that employing lysophosphatidylcholine except that 4% (w/v) bovine serum albumin, Fraction V (Sigma), was substituted for serum. The enzyme preparation to be assayed for hepatic lipase activity was preincubated at 0°C for 10 min in the presence of 1 M NaCl. All assays were carried out in a total volume of 200 ~1 at 37°C for 15 or 30 min in duplicate. One mU of enzyme activity represents the release of one nmol of fatty acid per min. For lipase inhibition assay, the enzyme source was mixed with increasing amounts of IgG fraction from immunized or non-immunized sera in a total volume of 100 ~1. The protein was adjusted with bovine serum albumin so that the final protein content was the same in all preincubation tubes. After preincuba-
217
tion at 0°C for 60 min, the remaining above.
lipase activity
was assayed as indicated
Heart perfusions
The hearts of rats fasted overnight were removed immediately after decapitation or while the animals were lightly anesthetized with diethyl ether. The hearts were perfused by the Langendorff method [3]. Except as specifically noted the perfusing media were not recirculated. Following an initial perfusion for a minimum of 2 min with Krebs-Ringer bicarbonate buffer (pH 7.4), the IgG fraction, from an immunized or non-immunized goat, dissolved in KrebsRinger buffer, or the buffer alone was recirculated through the heart for varying time periods. Immediately afterwards, the hearts were rinsed for 2 min by perfusion with Krebs-Ringer bicarbonate buffer. This was followed by perfusion with Krebs-Ringer buffer containing 1 unit heparin/ml. The heparin perfusate was collected and assayed for lipoprotein lipase activity. In experiments in which hearts were perfused with Krebs-Ringer buffer containing 14C-labelled chylomicrons, the chylomicron perfusate was collected in a single vessel containing 5 N NaOH and 14C02 production determined as previously described [12]. Collection and washing of labelled chylomicrons to decrease the amount of 14C-labelled unesterified fatty acids were carried out as previously described [ 31. Other procedures
Sodium dodecyl sulfate polyacrylamide gel electrophoresis was carried out according to Chen and Aladjem [13] and protein analysis was carried out by the method of Lowry et al. [ 141. Protein concentration was also determined by using an extinction coefficient of 13.1 at 280 nM for 1% IgG solution [15]. Double immunodiffusion was carried out on 1% agarose gel (w/v) in 0.05 M Tris * HCl, pH 7.4 [16]. Upon completion of the precipitation reaction the immunodiffusion plates were washed successively with 0.15 M NaCl and distilled water, dried and stained with 0.1% Amido Black A in 12% acetic acid (v/v) and 1.6% sodium acetate (w/v). Results Antibody
formation
Initially, in order to assess the immunologic response of the goat to purified rat heart lipoprotein lipase, serum was collected periodically and assayed for its ability to inhibit lipoprotein lipase activity. As shown in Fig. 1, anti-lipoprotein lipase activity could be detected in the goat serum within 20 days after the first injection; the titer increased somewhat during the following 50 days. A subsequent decline in the serum anti-lipoprotein lipase titer was reversed by further antigen injections. Antibody formation in the goat injected with purified lipoprotein lipase was confirmed by double immunodiffusion in agarose gel. Upon staining of the gel the IgG fraction isolated from goat antisera showed precipitin lines with concanavalin A and with rat heart lipoprotein lipase purified by heparin-Sepharose chromatography. When anti-concanavalin A reactivity was removed by passage
218
1OOr ,’ ,’
/’ I
0
60
120
180
‘1
240
PURIFIED HEART LPL
300
60
0
DAYS E‘IK:. I.
Immunr
rat
heart
thr
g<,at
protein tion,
l,‘ig. jnx
response
lipoprotriu xvas bled hpasr
2.
Mtathods
then
I:fftsct
amounts
(
-~~-~
stitutcd
with
rat
mU
for
anti-rat
heart
the
)
with
procedures IgG
fraction.
the
srrum
on
lipase (2
were
lipasr
activity lipasc
with LPL.
IgG
III hlethods.
lipase had
240
fraction
LPL.
activity
of
preincubated
bvcn
purified
by
in d barbital was from
drtcrmincd.
purifird
a time
at lvhich
of rat antisera
WHY
(1 mu)
with
wprrsents with
actiwty
lipoprotein
NBS injected Aliquots
fraction which
goat point
determined.
lipoprotein IgG
mLJ)
A
preincubated
as drscribetl
lipoprotein
sxnilar lipasr
fraction
activity
activltv.
BI’I’OWS. Each
activity)
activity
lipoprotrin
Iksidual
thv
titezl- of
lipoprotein lipaw
and
rc’prcscnt
IgG
vn/ymc by
lipasc lipasc
lipasc
of
indicatr~d lrpast,
lipoprotvin
heart
(A------A). anti-lipoprotein
inhibltlon
lipoprotein
lipoprotein
(D-----d)
, ___
by periods
anti-lipoprotc,in
0.84
assayed
goat
powder for
the to
at the
chromatography acctonc
and
(equal
1 : ROO),
monitored
lipasc
180
120 1gGCpg)
hrart
lipoprotein
rat
lipo-
(final
hcxart.
dilulipase.
Incrcas-
as describrd
in
heparin-St’pharosc extract Open
a non-immurlired
of
rat
heart
symbols goat
sub-
lipdw.
of this IgG fraction through a column of concanavalin A-Sepharose, the antilipoprotein lipase activity remaining upon staining was still found to show a line of precipitation with purified lipoprotein lipase but not with concanavalin A. Thus, the anti-concanavalin A reactivity was independent of the anti-lipoprotein lipase activity. Preincubation of a constant amount of purified rat heart lipoprotein lipase activity (2 mu) with increasing amounts of anti-lipoprotein lipase IgG fraction at 0°C for 1 h produced a progressive decrease of enzyme activity until approx. 98% inhibition was observed (Fig. 2). Preincubation with an IgG fraction prepared from serum of a non-immunized goat did not alter the original lipoprotein lipase activity (Fig. 2). In addition to the observed inhibition of activity of purified rat heart lipoprotein lipase, the anti-lipoprotein lipase IgG fraction similarly inhibited lipolytic activity in crude extracts of acetone powders of rat heart (Fig. 2). The crude extracts in Fig. 2 appear to be inhibited to a larger extent than is purified lipoprotein lipase activity; however, this is not the case, since the crude enzyme activity used was only half that employed for the experiments with the purified enzyme. Inhibition of enzyme activity by the goat anti-lipoprotein lipase IgG fraction occurred rapidly, the inhibition being greater than 80% after a lo-min preincubation at 0°C and essentially total inhibition was observed (98%) at 30 min (Fig. 3).
219
10
0
20
30
~~~C~ATl~~
40
TiME
(min
50
)
Fig. 3. Kinetics of inhibition of enzyme activity by anti-lipoprotein lipase IgG fraction. Aliquots of rat heart lipoprotein lipase activity (0.3 mU) purified bv hrparin-Sepharose chromatographv wverc prcincubated at 0°C for various times with 0.24 wg of either anti-lipoprotein lipasr Igti fraction (a--0) or IgG fraction from a non-immunized goat (i‘-). Residual lipoprotein lipase activity was dctvrmined. LPI,. lipoprotein lipaw.
The antibody prepared to purified rat heart lipoprotein lipase was tested for its capacity to inhibit the lipoprotein lipase activity derived from other tissues of the rat. After separation by heparin-Sepharose chromatography of the hepatic and extrahepatic lipases of rat postheparin plasma, these enzymes were preincubated with antibody to lipoprotein lipase and the residual lipase activity assayed (Fig. 4). The extrahepatic lipase was inhibited about 70%. by 30 ,ug of
.--c
o 100 80
60
EXTRAHEPATIC
HEPATIC
40
0
60
120
180
240
0
60
120
180
240
1gG fpg) Fig. 4. Effect of anti-rat heart fipoprotein lipase IgG fraction on the extra-hepatic and hepatic lipax activities of rat postheparin plasma. Plasma was obtained from fasting rats after heparin injection as described in the text and the plasma applied to a heparin-Sepharose column. Aliquots of the cluted fraction with the maximum lipolytic activity associated with hepatic and extra-hepatic lipases were preincubated with and of IgG fraction from serum of a increasing amounts of anti-lipoprotein lipase IgG fraction (a A) 0). The lipoprotein Iipase activity remaining was determined. LPL. lipopronon-immunized goat (o-tein iipase.
80
60 CRUDE
0
60
MAMMARY
120
180
240
IgG(j& Fig. 5. Effect of anti-lipoprotein lipase IgG fraction on lipoprotein lipase activity of rat adipose and mammary tissues. Increasing amounts of goat anti-rat heart lipoprotein lipase IgG fraction were weincubated as described in Methods with adipose tissue lipoprotein lipasr activitv (4.0 mu) which had been purified by heparinSepharose chromatography (a---•), and with adipose lipoprotein lipasc activity and nx~mmary lipoprotein lipase activity (1.0 mU) (m- --¤) in barbital extracts (1.2 mu) ( A--A) of acetone powders of thesr two latter tissues (see text). Residual lipoprotein lipase activity was deter-I-------O) represent similar procedures with serum IgG fracmined. Open symbols ( _-w, , -.---r\.. tion from a non-immunircd goat substituted for the anti-lipoprotein lipasc IgG fraction. LPL. lipoprotein lipasc.
the anti-lipoprotein lipase IgG fraction (Fig. 4), whereas the activity associated with the hepatic lipase was unaffected by the addition of the anti-lipoprotein lipase IgG fraction up to 240 pg (Fig. 4). Further, the anti-lipoprotein lipase IgG fraction totally inhibited the lipoprotein lipase activity partially purified from rat adipose tissue (Fig. 5) and, to a large extent, the lipoprotein lipase activity of crude extracts of acetone powders of rat adipose tissue and lactating rat mammary gland (Fig. 5). In marked contrast to these results, no inhibition of lipase activity by the anti-rat heart lipoprotein lipase IgG fraction could be demonstrated with partially purified heart lipoprotein lipase derived from three other species, rabbit, guinea pig and bovine; nor did the antibody inhibit lipoprotein lipase activity partially purified from bovine skim milk (Fig. 6).
Effect
of antibody
on lipoprokin
lipase activity
in the intact heart
Lipoprotein lipase activity extracted from several rat tissues was clearly shown to be inhibited by the antibody preparation. Whether this inhibition could also be shown in the intact tissue was studied using an isolated perfused heart preparation. Results in Table I show that when hearts were perfused with antibody and subsequently with heparin the lipoprotein lipase activity released by heparin was not greater than 4% of the lipase activity released by heparin from control hearts.
221
A.fX-MN
B.EmN
SKIM h%K
I-1EART
100 r-rr;?\
a0 60 0
60
D.GUW
m120
PIG
180
240
HEART
100
80
60 0
60
120
iao
240
Fig. 6. Effect
of anti-rat heart lipoprotein lipase IgG fraction on lipoprotein lipase activity of (A) bovine milk, (I3) bovine heart, (C) rabbit heart, and (D) guinea pig heart. Aliquots of lipoprotein lipase act.ivitY (4.8, 0.5, 1.7 and 0.2 mU, respectiveIy). which had been purified by heparin-Sepharose chromatography. were preincubated with increasing amounts of anti-lipoprotein lipase, IgG fraction f*----l). Residual 0) represents a similar procedure lipoprotein lipase activity was measured. The open symbol (“employing IgG fraction from a non-immunized goat. LPL, lipoprotein lipase.
The heparin perfusate from ailtibody-treated hearts was tested for its ability to inhibit the lipoprotein lipase activity in heparin perfusates from control hearts. Any residual capacity to inhibit could reflect the presence of free antibody. Data in Table II show that heparin-released lipoprotein lipase activity in perfusates from control hearts was unaffected by addition of the heparin perfusate from antibody-treated hearts. Clearly the antibody was bound to some site in the heart during perfusion. However, it was not determined whether the antibody was bound initially to lipoprotein lipase or whether the antibody-
TABLE EFFECT
PROTEIN
I
OF
PERFUSION
LIPASE
ACTIVITY
OF
ANTIBODIES BY HEPARIN
TO LIPOPROTEIN
LIPASE
ON RELEASE
OF LIPO-
Hearts of rats fasted overnight were perfused for 2 min with Krebs-Ringer bicarbonate buffer (pH 7.4) at 37OC at the rate of 6-8 mllmin. This buffer or immunized or non-immunized serum KgG was then recirculated through the heart for 5 min. 58 mg IgG was recirculated in a total volume of 20 mi buffer. A 2min. nonrecirculating perfusion with Krebs-Ringer was followed by perfusion with the same buffer containing 1 unit hcparin/ml. The heparin perfusate was collected in 2-ml aliquots and assayed in duplicate for lipoprotein lipase activity. Values shown are measurements of lipoprotein lipase activity (mU/ml) in the first Z-ml aliquot collected from individual hearts. The figures are means + SD.; the number of rats used is given in parentheses, ____IIExpt.
Buffer
No. 1
2 3
2.7 f 0.07 11.2 +_2.8 .,..-
(2) (2)
Non-immunized si?YUm fgG
Immunized serum 1gG
5.2 i 1.4 (4) 7.2 + 2.7 (4) 12.3 + 2.2 (3) -
0.14 I! 0.06 0.16 I 0.05 0.37 + 0.10
(4) (4) (3)
222
TABLE
II
TF:ST
F.OK
RliSlDUAL
TREATED Aliquots or
(0.05
ml)
of
non-immunized
without
the
lipaw.
thwc
heparin
serum
addition
lipoprotein of
INIIIBITOKY
ACTIVITY
IN
HEPAR1.X
PI
FROM
ANTIBODY-
IIEARTS
of
ml
given
control
from
(Controls.
0.05
Values
scparatc~
pcrfusatc
IgG
of
heparin
are
perfusatrs
the
prwiouslv
2.) wcw
perfusatr
average
and
fixr
of
perfused
assayed from
ivith
buffer
lipoprotein of
from
(C(>ntrol.
11pasr
prrviously
assavs
pcrfusates
+ hegarin
for
hearts
duplicate
srparatc
Control of
a heart
F:xp.
perfused
lipoprotein
with
Iipase
antibodv-treated
Exp.
acti\.ity
1)
xrith
and
antibodies
to
activitv
(mU/ml)
hearts.
prrfusate
antibody-treated
heart
1.8 2.0 2
4.2
4.4
4.1
4.5 4.2 .______..
____~__~__-_
_ _~_~
enzyme complex was formed in the perfusate after heparin administration. If the enzyme were, in fact, blocked by the antibody in situ, no hydrolysis of circulating triglyceride could occur. When hearts were perfused with [l-‘4C]palmitate-labelled chylomicrons, the mean values for pequiv. triacylglycerol fatty acid hydrolyzed per g heart, as determined by 14COZ production, were 0.32 for buffer control, 0.29 for IgG fraction from a non-immunized goat and 0.01 for antilipoprotein lipase antibody (Table III). Thus the effect of prior perfusion with antibody was to block by about 97% the hydrolysis of chylomicron triacylglycerol subsequently perfused through the hearts.
TABLE;
III
EYFECT
OF’
FUSED
ANTIBODIES
WIT11
TO
LIPOPROTEIN
1,IPASE
[I-‘4.CJPALMITATE-LABELI,ED
14C02
ON
PRODUCTION
CHYLOMICRON
BY
HEARTS
PER-
TRIACYLGLYCEROL
FATTY
ACID Ilearts
of
Then latrd of
rat
fasted
medium through
20
lating
ml
the
amount
“‘C-1abeIlrd through
the
“CO2
shown
was
free
fattv
acid
heart
and
oxidized
(pequiv.
immuniaad
fattv
serum
m the
Prepcrfusion
I&,
second
62 mg
column
present to
medium
wds
buffer
in
acid the
“C02s these
IgG
in
chvlr
min
arr
by
0.2% were
contribute
perfusion):
to
a 15.min
from
of
total up
formation
0.13; the
of 14C02
the
“C02
produced acid/g
per
(Hequiv. 15 min ____-__
Non-immunized Immunized
serum serum
* Corrected
IgG
Igti for
the
possible
0.45
0.32
=
0.45
0.29
*
0.14
0.01
*
contribution
of
14C-labelIcd
free
fatty
acid
in the
chylc.
same chylo-
column. If all the single
following serum
perfusion) Buffer
the The
produced,
obtained.
fatty
with
the
non-immunized total
volume
nonrecircu-
radioactivity.
during
min.
in a total
acids/ml). first
for
rccircu-
perfusion
in the
thr
7.4)
I&G was
A P-min
fatty
taken
the
buffer,
subtracted
recirculated
in 25 ml buffer.
are shown
was
medium
it could
was
(pf-I
serum
triacyglycerol
Thvs e values the
IgG
followed
prquiv.
buffcr
non-immunized
strum
was
perfusion
values
bicarbonate
or
recirculated
(60
’ 4C02.
for
fatty
heart/l5
When are
IgG
chylomicrons
assayed
Krebs-Ringer
serum
non-immuniycd
bicarbonate
free
acid/g 0.13.
of
serum
’ 4C-labelled
“C-labelled
with
immunur~d
immunized
Krcbs-Ringcxr
perfusate of
initially
buffer,
10 min. of
washed
containing
The
for mg
with
containing
micron
of
111
pwfused
either
heart
buffer;
perfusion
buffer
20 h wwc
containing
passage amounts
IgG.
0.16;
the
values
223
Discussion The evidence presented in this study supports the view that lipoprotein lipase(s) from various tissues of the rat are structurally similar. Thus, antibody produced against heart lipoprotein lipase was able to inhibit markedly the lipoprotein lipase activity of rat adipose and mammary tissues and of rat postheparin plasma, indicating that these lipoprotein lipases share common antigenie properties. It was surprising that the antibody to rat heart lipoprotein lipase did not inhibit lipoprotein lipase activity from heart tissue of other species, namely rabbit, bovine, and guinea pig. The antibody also had no effect on the lipoprotein lipase activity of bovine milk. In contrast, Hernell et al. [17] have reported that antibody to bovine milk lipoprotein lipase does inhibit both human postheparin plasma and human milk lipoprotein lipase activities. Apparently, bovine and human lipoprotein lipases share antigenic determinants, whereas rat enzyme is immunologically different from lipoprotein lipase of other species which we have tested. The recent publications of the amino acid compositions of lipoprotein lipase purified from bovine milk [18,19] and rat heart [6] allow comparisons with the amino acid content of rat postheparin plasma lipoprotein lipase previously reported by Fielding et al. [20]. The amino acid composition of the two enzymes isolated from rat tissues are quite similar. This similarity is consistent with the conclusion reached from our experiments with antibody to rat heart lipoprotein lipase, i.e., that lipoprotein lipases in various tissues of the rat are antigenically alike. Alternatively, the correspondence in amino acid composition between rat heart and postheparin plasma lipoprotein lipase may result from the heart’s being a source of the plasma activity. On the other hand, the amino acid composition of bovine milk lipoprotein lipase is markedly different from the rat lipoprotein lipase amino acid content. This distinct chemical difference probably accounts for the lack of inhibition by the rat antibody of lipoprotein lipase activity from other species. Recently Kompiang et al. [4] have demonstrated that injection of lipoprotein lipase antibody into chickens caused a marked increase in the plasma triglyceride concentration. These authors did not measure lipoprotein lipase activity in tissues but concluded that the enzyme was inhibited by the antibody within the plasma compartment. Our data support their conclusion since antibody to lipoprotein lipase was able to inhibit lipoprotein lipase activity in the intact isolated rat heart. It would appear that the antibody had access to the endothelial surface and that the lipoprotein lipase activity inhibited was that activity corresponding to the fraction readily released by heparin. This fraction is believed to be located close to the endothelial surface. It could not be ascertained from the experiments on release of lipoprotein lipase activity by heparin whether the antibody perfused through the heart formed a complex with the enzyme in the tissue and therefore prevented release of lipolytic activity into the perfusate or whether the enzyme * antibody complex was formed following release by heparin of the antibody and enzyme into the perfusate. However, addition of a second antibody, a rabbit antigoat IgG fraction, did show that the lipoprotein lipase antibody was released into the perfusate by heparin.
224
Our studies using [‘4C]chyle clearly distinguished between the two possibilities suggested above. The fact that a marked inhibition of hydrolysis was observed after perfusion of antibody strongly suggests that only the first possibility could be operative. The present study supports the conclusion that the functional fraction of lipoprotein lipase activity is on the endothelial surface. Final direct evidence for locating lipoprotein lipase will undoubtedly come from visualization experiments employing the electron microscope. Acknowledgments This work was supported in part by funds from: the National Institutes of Health, HL16577 and HL17246; the American Heart Association, 76-666; the American Heart Association, Greater Los Angeles Affiliate, 4921G4; and the Veterans Administration Medical Research. The authors wish to thank Judith Nikazy and Raymond Miller for their expert technical help. In addition, the authors would like to express their gratitude to Drs. I. Nilsson-Ehle and W. Palin for their help in the preparation of antiserum, and to Dr. R.O. SCOW for the gift of rat mammary gland acetone powder. References 1
Robinson,
D.S.
Elsevier,
Amsterdam
Garfinkel,
(1970)
A.S.
and
&hot,!, Robinson.
J.
Kompidng.
I.P.,
Twu.
Garfinkcl,
Chung,
.J. and
Garfinkel,
edn..
A.S. A.
and
Schotz,
Yang.
and
Schc)tL. J.
M.C.
Biol.
Cremer,
IIenson.
and
P. and
SchotL. Schotz,
M.C.
I’nser,
Kun~.
F.,
13.
Brown.
Res.
17.
E.II..
eds.).
51-l
22,
498-505
463-472
63-68 (1974) D.H.
.J. Biol.
(1970)
Chem.
Methods
249,
2220-2226
in Immunology.
13
Chen. Lowry,
C.-II.
and
Aladjem,
F.
15
Wcintraub,
16
Albers.
17
Hcrnell.
O.,
Egelrud,
T. and
18
Ivrrius,
P.-II.
and
Gstlund-Lindqvist,
19
Kinnunen.
20
Fielding.
O.J., Roscbrough. M.D. J.J.,
and
Chen,
P.K.J., P.E..
Biophys. Res.
Lipid
Opic,
Biochem.
Fan.
Schlamowitz,
C.H.
and
Huttunen.
Shore,
(1974)
N.J..
V.G.
A.L. M.
Aladjem,
L.H.
J.K. and
and
Acta 17,
and
409.
360.-366
536-541
Robinson,
D.S.
(1971)
Biophys.
Res.
Commun.
Randall,
R.J.
(1951)
Camp.
E‘. (1972)
Olivecrona,
T.
and
Biochem.
(1967)
Biochim.
(1976)
J.
Ehnholm, C.J.
C. (1974)
Biol. (1976)
60, ,J. Biol.
Physiol.
Biochemistry
(1975)
A.-M. Fielding,
2nd
Mass.
Biochem.
J.
104.
317 14
16.
17
.J. Lipid
Sussdorf,
J.
J..
111-l
W.V.
Biochim.
(1976)
Stat/,
press
Rrs.
and
dnd
274-278 11.
(1976)
in the
D. and
M.
Atherosclerosis
.J. Lipid
Reading,
Borcnvtajn,
M.W.
Chem.
(1975)
M.C.
8,
Res.
(1975)
N.E.
Benjamin.
(Florkin.
Rrs.
Lipid
M.C.
(1972)
J.S..
Nilsson-Ehle, M.B.,
Wang
Garvcy,
10 12
and
Steinberg,
11
Llpid J.
(1977)
W.A.
J.
(1970)
C.,
PP. 189-198, L.C.
(1967)
Ehnholm,
D.H.,
Biochemistry
D.S. A.
Scanu.
A.,
Campbell,
M.C.
Bensaduun,
AS.
Bensadoun,
Comprehensive
and
Borcns/tajn. *J.-S.,
in
11,
B. 38, Acta
251,
Biochim.
Biochemistry
554
Chem.
193.
265-275
513-522
57-63
Biophys. Chcm.
549~
381,
Biophys. 13,
233-241
7791-7795 4318-4323
Acta
450.
342-351
306-
Biochirnica ct Bioph.vsica @ Elsevier/North-Holland
Acla,
489 (1977)
Biomedical
211-22‘1
Press
BBA 570X.1
ANTIBODIES
TO LIPOPROTEIN
APPLICATION
TO PERFUSED
LIPASE HEART
MICHAEL ‘2. SCHOTZ, JER-SHUNG TWU, MARY E. PEDERSEN, .4RLENE S. GARFINKEL and JAYME BORENSZTAJN a Research L’etcrans Administration, Wadsworlh Horpital Department of Medicine, UCLA School of Medicine and Nutritional Sciences, School of Public Health, University 90024, and a Department of Pathology, Pritzkcr School Chicago, Ill. 60637 (IJ.S.A.)
(Received
CHI-HONG
CHEN,
Center, Los Angeles, Calif 90073, the Division of Environmental and of California, Los Angeles, Calif. of Medicine, University of Chicago,
May 31st, 1977)
Summary An antibody was prepared against purified rat heart lipoprotein lipase. 1. This antibody showed marked species specificity, It inhibited almost totally the lipoprotein lipase activity from all rat tissues examined (i.e., heart, adipose, postheparin plasma, and mammary gland), while having no effect on the activity of lipoprotein lipase partially purified from rabbit, guinea pig and bovine heart and from bovine milk. The antibody also had no effect on the hepatic lipase activity of rat postheparin plasma. 2. After antibody to rat heart lipoprotein lipase was recirculated for 5 min through isolated rat hearts, little or no lipoprotein lipase activity could be detected in the perfusate during O-20 s of a subsequent non-recirculating perfusion with buffer containing 1 unit heparin/ml. 3. Following recirculation of antibody to lipoprotein lipase for 10 min and a non-recirculating perfusion with buffer for 2 min, the hearts no longer oxidized any significant amounts of 14C-labelled palmitate chylomicron triacylglycerol fatty acid to 14C0, during a 15-min perfusion. The data give compelling evidence that the functional fraction of lipoprotein lipase in hearts is at the endothelial cell surface accessible to lipoprotein lipase antibody.
Introduction Lipoprotein lipase is the enzyme required for the uptake of circulating protein triglyceride fatty acid by various extrahepatic tissues [I]. This established concept is based on several major pieces of evidence. First,
lipowelllipo-
215
protein lipase activity in each tissue correlates directly with the uptake of triglyceride fatty acid by that tissue [ 21. Further, release of a fraction of lipoprotein lipase activity from a tissue prevents the uptake [3]. Finally, adminstration of antisera to lipoprotein lipase prevents the removal of lipoprotein triglyceride from the plasma compartment [ 41. Lipoprotein lipase activities from different tissues (heart and adipose, for example) share the common properties of inhibition by salt and activation by a specific serum apolipoprotein. However, the fact that these enzyme activities appear to be regulated by different mechanisms raises the question as to whether their protein structures are similar. Recently lipoprotein lipase has been highly purified from rat heart tissue [ 5,6]. Employing antibody produced against this purified heart enzyme, the present study indicates that lipoprotein lipases from various tissues of the rat are indeed immunologically similar. The effect of lipoprotein lipase antibody on the physiological function of the enzyme, that is, the hydrolysis of circulating lipoprotein triglyceride by the intact heart, has also been studied. Materials and Methods Enzyme sources Sprague-Dawley rats (150-250 g) maintained on Purina Laboratory Chow were used. The tissue sources were hearts of male rats fasted overnight, epididymal adipose tissue from male rats fed ad libitum, and mammary gland of female rats lactating for 13 days fed ad libitum. Acetone powders were prepared from these tissues [7] and extracted with 5 mM barbital buffer, pH 7.2 (designated barbital buffer) [ 51. Postheparin plasma was obtained by intravenous injection of heparin (Calbiochem), 20 units/100 g body weight, into male rats which had been fasted overnight and were anesthetized with diethyl ether. Two minutes after injection, the rats were decapitated, the blood collected and the plasma obtained stored at -20” C. Enzyme purification Lipoprotein lipase was purified from rat heart acetone powders as previously described [5], except that elution of the enzyme from the heparin-Sepharose column was accomplished in the presence of 20% glycerol (v/v). Glycerol has been found to stabilize the enzyme activity. This partially purified preparation was employed in most of the experiments described. For preparation of antiserum, the lipoprotein lipase obtained by heparin-Sepharose chromatography was further purified by affinity chromatography on concanavalin A-Sepharose (Pharmacia) columns (1 X 5 cm) following the procedure of Bensadoun et al. [S]. The enzyme was eluted with barbital buffer containing 1.2 M NaCl, 20% glycerol (v/v) and 0.6 M oc-methyl-D-mannoside. With this procedure 50-60% of the enzyme activity applied to the column was recovered. After dialysis at 4°C for 24 h against a large excess of barbital buffer containing 0.05 M NaCl, the enzyme was concentrated by vacuum dialysis. It was stored at -20°C and used within 10 days. This purified lipoprotein lipase fraction showed three protein-staining bands on 1% sodium dodecyl sulfate polyacrylamide gel electrophoresis, one with a mobility corresponding to an apparent molecular
216
weight of 68 000; the other two were identical to bands found in the eluate when no protein was applied to the concanavalin A column. These latter bands apparently resulted from leakage of the matrix-bound ligand, concanavalin A, during chromatography [S] . The hepatic and extrahepatic lipases from postheparin plasma were separated by heparin-Sepharose [5]. The plasma was diluted several-fold with 0.15 M NaCl. After centrifugation at 3000 X g for 30 min at 4°C the supernatant fraction was adjusted to 0.4 M NaCl and a 5-ml sample applied to a column of heparin-Sepharose (1.5 X 15 cm) equilibrated with barbital buffer containing 0.4 M NaCl. The column was washed with the same buffer. The hepatic lipase activity was eluted with barbital buffer containing 0.75 M NaCl and 20% glycerol (v/v) and the extrahepatic lipase activity with the same buffer containing 1.5 M NaCl. Recovery of the total triglyceride hydrolase activity was go-100%. Antibody production Approx. 15-20 pg of the purified heart enzyme, emulsified with an equal volume of Freund’s complete adjuvant, was injected subcutaneously at multiple sites into the suprascapular region of a 60-kg female goat. This procedure was repeated twice at 2-weekintervals, substituting incomplete for complete Freund’s adjuvant. Ten days after the last injection, the goat was bled from the jugular vein and the serum separated. The IgG fraction was then isolated by ammonium sulfate precipitation and DEAE-cellulose chromatography [9]. Upon 1% sodium dodecyl sulfate polyacrylamide gel electrophoresis, the IgG fraction thus obtained displayed two protein-staining bands, corresponding to the heavy and light chains of immunoglobulin. The IgG fraction (6.4 mg/ml) was stored at -20°C in 10 mM phosphate buffer (pH 7.4) containing 0.015 M NaCl and 0.02% NaN, (w/v). The IgG fraction of serum obtained from a non-immunized goat was isolated and stored in the same manner as described above. Anticoncanavalin A activity of the IgG fraction was removed by passing the IgG fraction through a concanavalin A-Sepharose column, the unbound fraction constituting the purified antibody to lipoprotein lipase. Lipase assays and incubation conditions Lipoprotein lipase activity was assayed using as substrate serum activated [9,10-3H]oleic acid-labelled trioleoylglycerol emulsified with lysophosphatidylcholine [lo] or phosphatidylcholine [ll]. The assay for hepatic lipase activity was the same as that employing lysophosphatidylcholine except that 4% (w/v) bovine serum albumin, Fraction V (Sigma), was substituted for serum. The enzyme preparation to be assayed for hepatic lipase activity was preincubated at 0°C for 10 min in the presence of 1 M NaCl. All assays were carried out in a total volume of 200 ~1 at 37°C for 15 or 30 min in duplicate. One mU of enzyme activity represents the release of one nmol of fatty acid per min. For lipase inhibition assay, the enzyme source was mixed with increasing amounts of IgG fraction from immunized or non-immunized sera in a total volume of 100 ~1. The protein was adjusted with bovine serum albumin so that the final protein content was the same in all preincubation tubes. After preincuba-
217
tion at 0°C for 60 min, the remaining above.
lipase activity
was assayed as indicated
Heart perfusions
The hearts of rats fasted overnight were removed immediately after decapitation or while the animals were lightly anesthetized with diethyl ether. The hearts were perfused by the Langendorff method [3]. Except as specifically noted the perfusing media were not recirculated. Following an initial perfusion for a minimum of 2 min with Krebs-Ringer bicarbonate buffer (pH 7.4), the IgG fraction, from an immunized or non-immunized goat, dissolved in KrebsRinger buffer, or the buffer alone was recirculated through the heart for varying time periods. Immediately afterwards, the hearts were rinsed for 2 min by perfusion with Krebs-Ringer bicarbonate buffer. This was followed by perfusion with Krebs-Ringer buffer containing 1 unit heparin/ml. The heparin perfusate was collected and assayed for lipoprotein lipase activity. In experiments in which hearts were perfused with Krebs-Ringer buffer containing 14C-labelled chylomicrons, the chylomicron perfusate was collected in a single vessel containing 5 N NaOH and 14C02 production determined as previously described [12]. Collection and washing of labelled chylomicrons to decrease the amount of 14C-labelled unesterified fatty acids were carried out as previously described [ 31. Other procedures
Sodium dodecyl sulfate polyacrylamide gel electrophoresis was carried out according to Chen and Aladjem [13] and protein analysis was carried out by the method of Lowry et al. [ 141. Protein concentration was also determined by using an extinction coefficient of 13.1 at 280 nM for 1% IgG solution [15]. Double immunodiffusion was carried out on 1% agarose gel (w/v) in 0.05 M Tris * HCl, pH 7.4 [16]. Upon completion of the precipitation reaction the immunodiffusion plates were washed successively with 0.15 M NaCl and distilled water, dried and stained with 0.1% Amido Black A in 12% acetic acid (v/v) and 1.6% sodium acetate (w/v). Results Antibody
formation
Initially, in order to assess the immunologic response of the goat to purified rat heart lipoprotein lipase, serum was collected periodically and assayed for its ability to inhibit lipoprotein lipase activity. As shown in Fig. 1, anti-lipoprotein lipase activity could be detected in the goat serum within 20 days after the first injection; the titer increased somewhat during the following 50 days. A subsequent decline in the serum anti-lipoprotein lipase titer was reversed by further antigen injections. Antibody formation in the goat injected with purified lipoprotein lipase was confirmed by double immunodiffusion in agarose gel. Upon staining of the gel the IgG fraction isolated from goat antisera showed precipitin lines with concanavalin A and with rat heart lipoprotein lipase purified by heparin-Sepharose chromatography. When anti-concanavalin A reactivity was removed by passage
218
1OOr ,’ ,’
/’ I
0
60
120
180
‘1
240
PURIFIED HEART LPL
300
60
0
DAYS E‘IK:. I.
Immunr
rat
heart
thr
g<,at
protein tion,
l,‘ig. jnx
response
lipoprotriu xvas bled hpasr
2.
Mtathods
then
I:fftsct
amounts
(
-~~-~
stitutcd
with
rat
mU
for
anti-rat
heart
the
)
with
procedures IgG
fraction.
the
srrum
on
lipase (2
were
lipasr
activity lipasc
with LPL.
IgG
III hlethods.
lipase had
240
fraction
LPL.
activity
of
preincubated
bvcn
purified
by
in d barbital was from
drtcrmincd.
purifird
a time
at lvhich
of rat antisera
WHY
(1 mu)
with
wprrsents with
actiwty
lipoprotein
NBS injected Aliquots
fraction which
goat point
determined.
lipoprotein IgG
mLJ)
A
preincubated
as drscribetl
lipoprotein
sxnilar lipasr
fraction
activity
activltv.
BI’I’OWS. Each
activity)
activity
lipoprotrin
Iksidual
thv
titezl- of
lipoprotein lipaw
and
rc’prcscnt
IgG
vn/ymc by
lipasc lipasc
lipasc
of
indicatr~d lrpast,
lipoprotvin
heart
(A------A). anti-lipoprotein
inhibltlon
lipoprotein
lipoprotein
(D-----d)
, ___
by periods
anti-lipoprotc,in
0.84
assayed
goat
powder for
the to
at the
chromatography acctonc
and
(equal
1 : ROO),
monitored
lipasc
180
120 1gGCpg)
hrart
lipoprotein
rat
lipo-
(final
hcxart.
dilulipase.
Incrcas-
as describrd
in
heparin-St’pharosc extract Open
a non-immurlired
of
rat
heart
symbols goat
sub-
lipdw.
of this IgG fraction through a column of concanavalin A-Sepharose, the antilipoprotein lipase activity remaining upon staining was still found to show a line of precipitation with purified lipoprotein lipase but not with concanavalin A. Thus, the anti-concanavalin A reactivity was independent of the anti-lipoprotein lipase activity. Preincubation of a constant amount of purified rat heart lipoprotein lipase activity (2 mu) with increasing amounts of anti-lipoprotein lipase IgG fraction at 0°C for 1 h produced a progressive decrease of enzyme activity until approx. 98% inhibition was observed (Fig. 2). Preincubation with an IgG fraction prepared from serum of a non-immunized goat did not alter the original lipoprotein lipase activity (Fig. 2). In addition to the observed inhibition of activity of purified rat heart lipoprotein lipase, the anti-lipoprotein lipase IgG fraction similarly inhibited lipolytic activity in crude extracts of acetone powders of rat heart (Fig. 2). The crude extracts in Fig. 2 appear to be inhibited to a larger extent than is purified lipoprotein lipase activity; however, this is not the case, since the crude enzyme activity used was only half that employed for the experiments with the purified enzyme. Inhibition of enzyme activity by the goat anti-lipoprotein lipase IgG fraction occurred rapidly, the inhibition being greater than 80% after a lo-min preincubation at 0°C and essentially total inhibition was observed (98%) at 30 min (Fig. 3).
219
10
0
20
30
~~~C~ATl~~
40
TiME
(min
50
)
Fig. 3. Kinetics of inhibition of enzyme activity by anti-lipoprotein lipase IgG fraction. Aliquots of rat heart lipoprotein lipase activity (0.3 mU) purified bv hrparin-Sepharose chromatographv wverc prcincubated at 0°C for various times with 0.24 wg of either anti-lipoprotein lipasr Igti fraction (a--0) or IgG fraction from a non-immunized goat (i‘-). Residual lipoprotein lipase activity was dctvrmined. LPI,. lipoprotein lipaw.
The antibody prepared to purified rat heart lipoprotein lipase was tested for its capacity to inhibit the lipoprotein lipase activity derived from other tissues of the rat. After separation by heparin-Sepharose chromatography of the hepatic and extrahepatic lipases of rat postheparin plasma, these enzymes were preincubated with antibody to lipoprotein lipase and the residual lipase activity assayed (Fig. 4). The extrahepatic lipase was inhibited about 70%. by 30 ,ug of
.--c
o 100 80
60
EXTRAHEPATIC
HEPATIC
40
0
60
120
180
240
0
60
120
180
240
1gG fpg) Fig. 4. Effect of anti-rat heart fipoprotein lipase IgG fraction on the extra-hepatic and hepatic lipax activities of rat postheparin plasma. Plasma was obtained from fasting rats after heparin injection as described in the text and the plasma applied to a heparin-Sepharose column. Aliquots of the cluted fraction with the maximum lipolytic activity associated with hepatic and extra-hepatic lipases were preincubated with and of IgG fraction from serum of a increasing amounts of anti-lipoprotein lipase IgG fraction (a A) 0). The lipoprotein Iipase activity remaining was determined. LPL. lipopronon-immunized goat (o-tein iipase.
80
60 CRUDE
0
60
MAMMARY
120
180
240
IgG(j& Fig. 5. Effect of anti-lipoprotein lipase IgG fraction on lipoprotein lipase activity of rat adipose and mammary tissues. Increasing amounts of goat anti-rat heart lipoprotein lipase IgG fraction were weincubated as described in Methods with adipose tissue lipoprotein lipasr activitv (4.0 mu) which had been purified by heparinSepharose chromatography (a---•), and with adipose lipoprotein lipasc activity and nx~mmary lipoprotein lipase activity (1.0 mU) (m- --¤) in barbital extracts (1.2 mu) ( A--A) of acetone powders of thesr two latter tissues (see text). Residual lipoprotein lipase activity was deter-I-------O) represent similar procedures with serum IgG fracmined. Open symbols ( _-w, , -.---r\.. tion from a non-immunircd goat substituted for the anti-lipoprotein lipasc IgG fraction. LPL. lipoprotein lipasc.
the anti-lipoprotein lipase IgG fraction (Fig. 4), whereas the activity associated with the hepatic lipase was unaffected by the addition of the anti-lipoprotein lipase IgG fraction up to 240 pg (Fig. 4). Further, the anti-lipoprotein lipase IgG fraction totally inhibited the lipoprotein lipase activity partially purified from rat adipose tissue (Fig. 5) and, to a large extent, the lipoprotein lipase activity of crude extracts of acetone powders of rat adipose tissue and lactating rat mammary gland (Fig. 5). In marked contrast to these results, no inhibition of lipase activity by the anti-rat heart lipoprotein lipase IgG fraction could be demonstrated with partially purified heart lipoprotein lipase derived from three other species, rabbit, guinea pig and bovine; nor did the antibody inhibit lipoprotein lipase activity partially purified from bovine skim milk (Fig. 6).
Effect
of antibody
on lipoprokin
lipase activity
in the intact heart
Lipoprotein lipase activity extracted from several rat tissues was clearly shown to be inhibited by the antibody preparation. Whether this inhibition could also be shown in the intact tissue was studied using an isolated perfused heart preparation. Results in Table I show that when hearts were perfused with antibody and subsequently with heparin the lipoprotein lipase activity released by heparin was not greater than 4% of the lipase activity released by heparin from control hearts.
221
A.fX-MN
B.EmN
SKIM h%K
I-1EART
100 r-rr;?\
a0 60 0
60
D.GUW
m120
PIG
180
240
HEART
100
80
60 0
60
120
iao
240
Fig. 6. Effect
of anti-rat heart lipoprotein lipase IgG fraction on lipoprotein lipase activity of (A) bovine milk, (I3) bovine heart, (C) rabbit heart, and (D) guinea pig heart. Aliquots of lipoprotein lipase act.ivitY (4.8, 0.5, 1.7 and 0.2 mU, respectiveIy). which had been purified by heparin-Sepharose chromatography. were preincubated with increasing amounts of anti-lipoprotein lipase, IgG fraction f*----l). Residual 0) represents a similar procedure lipoprotein lipase activity was measured. The open symbol (“employing IgG fraction from a non-immunized goat. LPL, lipoprotein lipase.
The heparin perfusate from ailtibody-treated hearts was tested for its ability to inhibit the lipoprotein lipase activity in heparin perfusates from control hearts. Any residual capacity to inhibit could reflect the presence of free antibody. Data in Table II show that heparin-released lipoprotein lipase activity in perfusates from control hearts was unaffected by addition of the heparin perfusate from antibody-treated hearts. Clearly the antibody was bound to some site in the heart during perfusion. However, it was not determined whether the antibody was bound initially to lipoprotein lipase or whether the antibody-
TABLE EFFECT
PROTEIN
I
OF
PERFUSION
LIPASE
ACTIVITY
OF
ANTIBODIES BY HEPARIN
TO LIPOPROTEIN
LIPASE
ON RELEASE
OF LIPO-
Hearts of rats fasted overnight were perfused for 2 min with Krebs-Ringer bicarbonate buffer (pH 7.4) at 37OC at the rate of 6-8 mllmin. This buffer or immunized or non-immunized serum KgG was then recirculated through the heart for 5 min. 58 mg IgG was recirculated in a total volume of 20 mi buffer. A 2min. nonrecirculating perfusion with Krebs-Ringer was followed by perfusion with the same buffer containing 1 unit hcparin/ml. The heparin perfusate was collected in 2-ml aliquots and assayed in duplicate for lipoprotein lipase activity. Values shown are measurements of lipoprotein lipase activity (mU/ml) in the first Z-ml aliquot collected from individual hearts. The figures are means + SD.; the number of rats used is given in parentheses, ____IIExpt.
Buffer
No. 1
2 3
2.7 f 0.07 11.2 +_2.8 .,..-
(2) (2)
Non-immunized si?YUm fgG
Immunized serum 1gG
5.2 i 1.4 (4) 7.2 + 2.7 (4) 12.3 + 2.2 (3) -
0.14 I! 0.06 0.16 I 0.05 0.37 + 0.10
(4) (4) (3)
222
TABLE
II
TF:ST
F.OK
RliSlDUAL
TREATED Aliquots or
(0.05
ml)
of
non-immunized
without
the
lipaw.
thwc
heparin
serum
addition
lipoprotein of
INIIIBITOKY
ACTIVITY
IN
HEPAR1.X
PI
FROM
ANTIBODY-
IIEARTS
of
ml
given
control
from
(Controls.
0.05
Values
scparatc~
pcrfusatc
IgG
of
heparin
are
perfusatrs
the
prwiouslv
2.) wcw
perfusatr
average
and
fixr
of
perfused
assayed from
ivith
buffer
lipoprotein of
from
(C(>ntrol.
11pasr
prrviously
assavs
pcrfusates
+ hegarin
for
hearts
duplicate
srparatc
Control of
a heart
F:xp.
perfused
lipoprotein
with
Iipase
antibodv-treated
Exp.
acti\.ity
1)
xrith
and
antibodies
to
activitv
(mU/ml)
hearts.
prrfusate
antibody-treated
heart
1.8 2.0 2
4.2
4.4
4.1
4.5 4.2 .______..
____~__~__-_
_ _~_~
enzyme complex was formed in the perfusate after heparin administration. If the enzyme were, in fact, blocked by the antibody in situ, no hydrolysis of circulating triglyceride could occur. When hearts were perfused with [l-‘4C]palmitate-labelled chylomicrons, the mean values for pequiv. triacylglycerol fatty acid hydrolyzed per g heart, as determined by 14COZ production, were 0.32 for buffer control, 0.29 for IgG fraction from a non-immunized goat and 0.01 for antilipoprotein lipase antibody (Table III). Thus the effect of prior perfusion with antibody was to block by about 97% the hydrolysis of chylomicron triacylglycerol subsequently perfused through the hearts.
TABLE;
III
EYFECT
OF’
FUSED
ANTIBODIES
WIT11
TO
LIPOPROTEIN
1,IPASE
[I-‘4.CJPALMITATE-LABELI,ED
14C02
ON
PRODUCTION
CHYLOMICRON
BY
HEARTS
PER-
TRIACYLGLYCEROL
FATTY
ACID Ilearts
of
Then latrd of
rat
fasted
medium through
20
lating
ml
the
amount
“‘C-1abeIlrd through
the
“CO2
shown
was
free
fattv
acid
heart
and
oxidized
(pequiv.
immuniaad
fattv
serum
m the
Prepcrfusion
I&,
second
62 mg
column
present to
medium
wds
buffer
in
acid the
“C02s these
IgG
in
chvlr
min
arr
by
0.2% were
contribute
perfusion):
to
a 15.min
from
of
total up
formation
0.13; the
of 14C02
the
“C02
produced acid/g
per
(Hequiv. 15 min ____-__
Non-immunized Immunized
serum serum
* Corrected
IgG
Igti for
the
possible
0.45
0.32
=
0.45
0.29
*
0.14
0.01
*
contribution
of
14C-labelIcd
free
fatty
acid
in the
chylc.
same chylo-
column. If all the single
following serum
perfusion) Buffer
the The
produced,
obtained.
fatty
with
the
non-immunized total
volume
nonrecircu-
radioactivity.
during
min.
in a total
acids/ml). first
for
rccircu-
perfusion
in the
thr
7.4)
I&G was
A P-min
fatty
taken
the
buffer,
subtracted
recirculated
in 25 ml buffer.
are shown
was
medium
it could
was
(pf-I
serum
triacyglycerol
Thvs e values the
IgG
followed
prquiv.
buffcr
non-immunized
strum
was
perfusion
values
bicarbonate
or
recirculated
(60
’ 4C02.
for
fatty
heart/l5
When are
IgG
chylomicrons
assayed
Krebs-Ringer
serum
non-immuniycd
bicarbonate
free
acid/g 0.13.
of
serum
’ 4C-labelled
“C-labelled
with
immunur~d
immunized
Krcbs-Ringcxr
perfusate of
initially
buffer,
10 min. of
washed
containing
The
for mg
with
containing
micron
of
111
pwfused
either
heart
buffer;
perfusion
buffer
20 h wwc
containing
passage amounts
IgG.
0.16;
the
values
223
Discussion The evidence presented in this study supports the view that lipoprotein lipase(s) from various tissues of the rat are structurally similar. Thus, antibody produced against heart lipoprotein lipase was able to inhibit markedly the lipoprotein lipase activity of rat adipose and mammary tissues and of rat postheparin plasma, indicating that these lipoprotein lipases share common antigenie properties. It was surprising that the antibody to rat heart lipoprotein lipase did not inhibit lipoprotein lipase activity from heart tissue of other species, namely rabbit, bovine, and guinea pig. The antibody also had no effect on the lipoprotein lipase activity of bovine milk. In contrast, Hernell et al. [17] have reported that antibody to bovine milk lipoprotein lipase does inhibit both human postheparin plasma and human milk lipoprotein lipase activities. Apparently, bovine and human lipoprotein lipases share antigenic determinants, whereas rat enzyme is immunologically different from lipoprotein lipase of other species which we have tested. The recent publications of the amino acid compositions of lipoprotein lipase purified from bovine milk [18,19] and rat heart [6] allow comparisons with the amino acid content of rat postheparin plasma lipoprotein lipase previously reported by Fielding et al. [20]. The amino acid composition of the two enzymes isolated from rat tissues are quite similar. This similarity is consistent with the conclusion reached from our experiments with antibody to rat heart lipoprotein lipase, i.e., that lipoprotein lipases in various tissues of the rat are antigenically alike. Alternatively, the correspondence in amino acid composition between rat heart and postheparin plasma lipoprotein lipase may result from the heart’s being a source of the plasma activity. On the other hand, the amino acid composition of bovine milk lipoprotein lipase is markedly different from the rat lipoprotein lipase amino acid content. This distinct chemical difference probably accounts for the lack of inhibition by the rat antibody of lipoprotein lipase activity from other species. Recently Kompiang et al. [4] have demonstrated that injection of lipoprotein lipase antibody into chickens caused a marked increase in the plasma triglyceride concentration. These authors did not measure lipoprotein lipase activity in tissues but concluded that the enzyme was inhibited by the antibody within the plasma compartment. Our data support their conclusion since antibody to lipoprotein lipase was able to inhibit lipoprotein lipase activity in the intact isolated rat heart. It would appear that the antibody had access to the endothelial surface and that the lipoprotein lipase activity inhibited was that activity corresponding to the fraction readily released by heparin. This fraction is believed to be located close to the endothelial surface. It could not be ascertained from the experiments on release of lipoprotein lipase activity by heparin whether the antibody perfused through the heart formed a complex with the enzyme in the tissue and therefore prevented release of lipolytic activity into the perfusate or whether the enzyme * antibody complex was formed following release by heparin of the antibody and enzyme into the perfusate. However, addition of a second antibody, a rabbit antigoat IgG fraction, did show that the lipoprotein lipase antibody was released into the perfusate by heparin.
224
Our studies using [‘4C]chyle clearly distinguished between the two possibilities suggested above. The fact that a marked inhibition of hydrolysis was observed after perfusion of antibody strongly suggests that only the first possibility could be operative. The present study supports the conclusion that the functional fraction of lipoprotein lipase activity is on the endothelial surface. Final direct evidence for locating lipoprotein lipase will undoubtedly come from visualization experiments employing the electron microscope. Acknowledgments This work was supported in part by funds from: the National Institutes of Health, HL16577 and HL17246; the American Heart Association, 76-666; the American Heart Association, Greater Los Angeles Affiliate, 4921G4; and the Veterans Administration Medical Research. The authors wish to thank Judith Nikazy and Raymond Miller for their expert technical help. In addition, the authors would like to express their gratitude to Drs. I. Nilsson-Ehle and W. Palin for their help in the preparation of antiserum, and to Dr. R.O. SCOW for the gift of rat mammary gland acetone powder. References 1
Robinson,
D.S.
Elsevier,
Amsterdam
Garfinkel,
(1970)
A.S.
and
&hot,!, Robinson.
J.
Kompidng.
I.P.,
Twu.
Garfinkcl,
Chung,
.J. and
Garfinkel,
edn..
A.S. A.
and
Schotz,
Yang.
and
Schc)tL. J.
M.C.
Biol.
Cremer,
IIenson.
and
P. and
SchotL. Schotz,
M.C.
I’nser,
Kun~.
F.,
13.
Brown.
Res.
17.
E.II..
eds.).
51-l
22,
498-505
463-472
63-68 (1974) D.H.
.J. Biol.
(1970)
Chem.
Methods
249,
2220-2226
in Immunology.
13
Chen. Lowry,
C.-II.
and
Aladjem,
F.
15
Wcintraub,
16
Albers.
17
Hcrnell.
O.,
Egelrud,
T. and
18
Ivrrius,
P.-II.
and
Gstlund-Lindqvist,
19
Kinnunen.
20
Fielding.
O.J., Roscbrough. M.D. J.J.,
and
Chen,
P.K.J., P.E..
Biophys. Res.
Lipid
Opic,
Biochem.
Fan.
Schlamowitz,
C.H.
and
Huttunen.
Shore,
(1974)
N.J..
V.G.
A.L. M.
Aladjem,
L.H.
J.K. and
and
Acta 17,
and
409.
360.-366
536-541
Robinson,
D.S.
(1971)
Biophys.
Res.
Commun.
Randall,
R.J.
(1951)
Camp.
E‘. (1972)
Olivecrona,
T.
and
Biochem.
(1967)
Biochim.
(1976)
J.
Ehnholm, C.J.
C. (1974)
Biol. (1976)
60, ,J. Biol.
Physiol.
Biochemistry
(1975)
A.-M. Fielding,
2nd
Mass.
Biochem.
J.
104.
317 14
16.
17
.J. Lipid
Sussdorf,
J.
J..
111-l
W.V.
Biochim.
(1976)
Stat/,
press
Rrs.
and
dnd
274-278 11.
(1976)
in the
D. and
M.
Atherosclerosis
.J. Lipid
Reading,
Borcnvtajn,
M.W.
Chem.
(1975)
M.C.
8,
Res.
(1975)
N.E.
Benjamin.
(Florkin.
Rrs.
Lipid
M.C.
(1972)
J.S..
Nilsson-Ehle, M.B.,
Wang
Garvcy,
10 12
and
Steinberg,
11
Llpid J.
(1977)
W.A.
J.
(1970)
C.,
PP. 189-198, L.C.
(1967)
Ehnholm,
D.H.,
Biochemistry
D.S. A.
Scanu.
A.,
Campbell,
M.C.
Bensaduun,
AS.
Bensadoun,
Comprehensive
and
Borcns/tajn. *J.-S.,
in
11,
B. 38, Acta
251,
Biochim.
Biochemistry
554
Chem.
193.
265-275
513-522
57-63
Biophys. Chcm.
549~
381,
Biophys. 13,
233-241
7791-7795 4318-4323
Acta
450.
342-351
306-