Download Human Hepatic d-Aminolaevulinate Synthase

Clinical Science and Molecular Medicine (1977) 52, 509-521.
Human hepatic 6-aminolaevulinate synthase: requirement of
an exogenous system for succinyl-coenzyme A generation to
demonstrate increased activity in cirrhotic and
anticonvulsant-treated subjects
H. L. BONKOWSKY A N D JOANNE S. POMEROY
Hepatology Research Laboratory of Veterans' Administration Center, White River Junction, VT. and
Dartmouth Medical School, Hanover, NH, U.S.A.
(Received 19 July 1976; accepted 30 December 1976)
summary
1. We have studied activity of baminolaevulinate synthase in needle liver biopsy specimens
obtained from 12 human cirrhotic subjects, five
subjects who had ingested anticonvulsants and
from control subjects. Liver iron concentrations
and quantitative urinary excretions of porphyrins plus their precursors were also determined.
2. In liver homogenates from subjects of each
group, addition of an exogenous system for
generation of succinyl-coenzyme A (&A), including succinic thiokinase, resulted in appreciable enhancement of activity beyond that
obtained without this system.
3. Mean activities for 6-aminolaevulinate
synthase were not significantly different among
patient groups when assayed without the exogenous succinyl-CoA-generating system, but
liver homogenates from cirrhotic patients and
subjects ingesting anticonvulsants had significantly higher activities than control subjects in
the presence of the succinyl-CoA-generating
system.
4. Although mean liver iron concentration in
the cirrhotic group was slightly higher than in
control subjects, and in control subjects there
was some correlation between liver iron concentration and activity of 6-aminolaevulinate
synthase, variations in this activity could not
be accounted for solely on the basis of chronic
hepatic deposition. Nor were these variations
ascribable to differences among subjects in
ingestion of ethanol before biopsy or severity
Correspondence: Dr Herbert L. Bonkowsky, VA
Center, White River Junction, VT 05001, USA.
509
of hepatic inflammation as judged biochemically
and histologically.
5. Cirrhotic subjects excreted more uro- and
copro-porphyrin than control subjects, whereas
subjects ingesting anticonvulsantsexcreted more
6-aminolaevulinic acid and porphobilinogen
than control subjects. However, these increases
were small and not sufficient to account for all
the increased 6-aminolaevulinic acid which
potentially could have been formed by these
subjects.
6. These considerations raise the possibilities
that, in vivo: (a) rate of human hepatic synthesis
of 6-aminolaevulinic acid is modulated by the
supply of succinyl-CoA; (b) the rate of hepatic
synthesis of haem is increased in cirrhotic
patients and subjects ingesting anticonvulsants;
(c) other important routes exist for disposition
of haem precursors in these subjects, besides
utilization for haem synthesis.
Key words : alcohol, 8-aminolaevulinic acid
synthase, anticonvulsants, iron, liver cirrhosis,
porphobilinogen, porphyrins, succinyl-CoA.
Abbreviations:ALA, 6-aminolaevulinate; CoA,
coenzyme A; GSH, reduced glutathione.
Introduction
Control of hepatic haem synthesis is vested
importantly in the fist step, catalysed by the
intramitochondrial enzyme 8-aminolaevulinate
(ALA) synthase(succinyl-CoA-glycine succinyltransferase, EC 2.3.1.37) (Tschudy, 1974).
510
H . L. Bonkowsky and Joanne S. Pomeroy
Activity of this enzyme in liver extracts is subject to wide variation, generally believed to relate
primarily to the number of enzyme molecules
present, since this enzyme has a short half-life
and can be induced or repressed by a variety of
compounds. Nevertheless there are other
possible controlling mechanisms which may
modulate its activity, for example, availability
of the substrates glycine and succinyl-CoA.
Attainment of maximal ALA synthase activity
in uitro requires high concentrations of glycine
(Tschudy, Welland, Collins & Hunter, 1964;
Marver, Tschudy, Perlroth & Collins, 1966),
and there is evidence that depletion of the
glycine pool in uiuo can diminish activity of the
enzyme in porphyric rats (Piper, Condie &
Tephly, 1973). In contrast, when considering
liver homogenate, even from such rats, it has
been thought that supplies of succinyl-CoA,
generated endogenously, are adequate to supply
optimum concentrations of this substrate
(Marver et af., 1966). It has not been clear
whether this is also true for human liver
homogenates.
The role of hepatic iron deposition in control
of activity of ALA synthase and haem synthesis
also is understood imperfectly. Previous studies
have concentrated on iron as an inhibitor of
uroporphyrinogen 111 formation (Kushner, Lee
& Nacht, 1972) and utilization (Kushner,
Steinmuller & Lee, 1975), but large doses of
ferric citrate, given acutely to rats, exerted a
remarkable synergistic effect on drug-mediated
induction of hepatic ALA synthase (Stein,
Tschudy, Corcoran & Collins, 1970). Because
enhanced hepatic iron deposition occurs with
increased frequency in cirrhosis (Conn, 1975),
this observation is of potential importance in
understanding the abnormalities in porphyrin
metabolism which are common in liver disease,
especially cirrhosis, ranging from mild increases
in urinary porphyrin excretion to overt cutaneous porphyria (Doss, Look, Henning
Nawrocki, Schmidt, Dolle, Korb, Liiders &
Strohmeyer, 1972).
In one previous study (Levere, 1967), activity
of ALA synthase was found to be increased in
mitochondria isolated from human cirrhotic
livers when incubated with exogenously added
glycine and citrate, the latter to provide a source
for succinyl-CoA generated by activity of the
Krebs cycle. Interpretation of these results is
complicated by the fact that the livers studied
were obtained at surgery under anaesthesia induced by thiopentane, one of a group of drugs
well known to induce hepatic ALA synthase
(Tschudy, 1974; Granick, 1966). Also, the
cirrhotic patients had significant degrees of
portal hypertension, which may increase hepatic
iron deposition beyond that associated with
cirrhosis alone (Conn, 1972); thus perhaps it
was hepatic siderosis, rather than cirrhosis per
se, which led to increased activity of ALA
synthase. Finally, since ethanol has been found
to induce ALA synthase in rat liver (Shanley,
Zail & Joubert, 1968; Rubin, Bacchin, Gang &
Lieber, 1970), it was possible that excessive ingestion of ethanol explained the increased
activity of hepatic ALA synthase observed in
cirrhotic subjects.
To resolve some of these uncertainties and to
learn more about the control of human hepatic
ALA, porphyrin and haem synthesis in health
and disease, we have carried out the studies
herein described.
Methods
Subjects
With one exception (subject no. 6, in whom
only a surgical wedge biopsy was done) all
subjects underwent aspiration needle biopsy of
the liver for diagnostic or prognostic purposes.
Before biopsy the patients consented to having
their urine collected and agreed that, if liver
tissue was obtained at biopsy beyond that required for histopathological interpretation (core
more than 3 cm in length), this extra tissue could
be used for special studies.
In subjects with normal tests of liver function
who underwent needle liver biopsy, the procedure was performed to aid in the diagnosis of
fever of unknown origin, to search for suspected
infiltrative disease of the liver, or to assess degree
of fibrosis after previous methotrexate therapy
for psoriasis. In all instances, biopsies were done
without complication after at least 12 h of fasting, without premedication. Subject no. 6 underwent open surgical biopsy at the time of a
gastric fundoplication for reflux oesophagitis.
Subject no. 28 underwent a needle biopsy and
later an open surgical biopsy, the latter during a
cholecystectomy. In both instances informed,
written consent was obtained for wedge biopsy
before surgery. Ether induction and maintenance
Human hepatic ALA synthase
anaesthesia was used in both operations; no
barbiturates were used. Except as noted, the
subjects were not on medication before biopsy.
These protocols were approved by the Committee for Protection of Human Subjects of the
Dartmouth-Hitchock Medical Center.
Liver size was estimated by percussion, and
where possible by palpation, in the right midclavicular and mid-sternal lines. Determinations
of serum iron, total iron-binding capacity,
standard tests of liver function and urinary
creatinine concentrations were performed
by the clinical laboratory.
Twenty-four-hour urine collections, kept dark
and refrigerated, were obtained within 1 or 2
days of the time of liver biopsies and analyses
run on fresh urine or urine frozen (-20°C) not
longer than 2 weeks.
Liver biopsies and histology
Needle biopsies of liver were obtained by
aspiration, and portions for histological study
fixed in Carnoy’s solution and stained with
Masson’s Trichrome, Haematoxylin and Eosin
and Perl’s stains. Other portions were bathed in
iced NaCl solution (150 mmol/l) in iron-free
vials, taken to the laboratory, and portions used
for quantitative iron analysis and assay of ALA
synthase activity. Assay of this enzyme was
carried out on homogenates of fresh liver or on
homogenates of liver frozen for 1-42 days at
- 85°C (average duration of frozen storage, 9
days).
Histological examination of biopsy specimens
was carried out, before biochemical studies had
been completed, independently by a pathologist
and by the senior investigator. A semi-quantitative grading system was used by both observers, who were found always to agree exactly on
the degree of fibrosis and to agree or be only one
grade apart on degree of necrosis, and fat or iron
deposition. The grading systems were as follows.
Degree of fibrosis: 0, no increase; 1 + , any
increase but no cirrhosis (this indicated some
stellate portal fibrosis or central hyaline
sclerosis);2 +, a cirrhotic pattern but with rather
slender fibrous septa; 3 + ,moderately advanced
cirrhosis with thick septa; 4 +, extremely advanced cirrhosis (very thick septa with only
small islands of regenerating hepatocytes remaining).
Degree of necrosis: 0, none; 1 +, minimal
511
evidenceof necrosis of liver cells as evidenced by
presence of rare acidophilic bodies or balloon
cells (with or without Mallory bodies), affecting
< 5 % of cells ;2 ,moderate evidence of necrosis, affecting from about 5% to one-third of
cells; 3 + ,severe necrosis, affecting one- to twothirds of liver cells; 4 + , very severe necrosis,
affecting more than two-thirds of cells.
Degree of fatty change: 0, none; 1 + ,minimal
fatty change, affecting <S% of hepatocytes;
2 + ,moderate fatty change, affectingfrom about
5% to one-third of cells; 3 +, severefatty change,
affecting from one- to two-thirds of cells; 4 + ,
very severe fatty change, affecting more than
two-thirds of cells. Degree of iron deposition
was graded by the criteria of Scheuer, Williams
& Muir (1962).
+
Enzyme assays
Homogenates of 3-20 mg wet weight of liver,
prepared with a micro-homogenizer (MicroMetrics, Cleveland, Ohio, U.S.A.), were assayed
for activity of ALA synthase by the method of
Ebert, Tschudy, Chowdhry & Chirigos (1970),
a final concentration of Tris chloride (PH 7.4) of
150 mmol/l being used. lpCi of [2,3-14C]succinic
acid (New England Nuclear Corp., 12.6 mCi/
mrnol) was added to start the reaction and provide labelled substrate.
When sufficient tissue was available, ALA
synthase activity was measured in homogenates
both in the presenceand absenceof anexogenous
system for generation of succinyl-CoA. This
system was composed of MgCL, 5 mmol/l,
succinic thiokinase (succinateCoA ligase
[ADP], EC 6.2.1.5), 0.35-040 unitlml, and
freshly prepared solutions of ATP, 5 mrnol/l,
GSH, 5 mmol/l, and CoA, 50 pmol/l (final concentrations). Preliminary experiments had indicated that these concentrations of succinic thiokinase led to maximal rates of ALA formation.
The succinic thiokinase added was usually
contained in a 105 OOO g supernatant from sonicated H-5 mutant strain of Rhodopseudomonas
spheroides, which is rich in this enzyme but
deficientin ALA synthase. In a few experiments,
succinic thiokinase from Porphyrin Products
(Logan, Utah U.S.A.) was used; it gave comparable results but led to higher radioactivity
counts in the reagent control systems.
In all experiments, suitable reagent controls
were included, containing everythingexcept liver
H . L. Bonkowsky and Joanne S. Pomeroy
512
homogenate, and results were corrected for them
and for recovery of ALA (80%) determined with
[5-I4C]ALA (New England Nuclear, 26.2
mCi/mmol). Addition of sodium fluoride (50
mmol/l) to inhibit adenosine triphosphatases
(Yoda, Schachter & Israels, 1975), or of pyruvate kinase (ATP-pyruvate phosphotransferase,
EC 2.7.1.40), ADP and phosphoenol pyruvate
to generate ATP continuously (Patton & Beattie,
1973) did not enhance formation of ALA beyond
that obtainable with addition of succinic thiokinase, ATP, MgCL, GSH and CoA, as outlined above. In addition, some homogenates of
fresh liver were frozen-and-thawed three times
in methanol/solid carbon dioxide before assay
since this has been reported to increase markedly
apparent activity of ALA synthase in isolated
rat liver mitochondria (Patton & Beattie, 1973).
Succinic thiokinase activity was measured by
the method of Kaufman, Gilvarg, Cori & Ochoa
(1953).
Other methods
Tissue iron content was determined by the
method of Barry & Sherlock (1971), with a
coefficient of variation of 8% and average recovery of iron of 102%. Urinary ALA and porphobilinogen were estimated by the method of
Mauzerall & Granick (1956), and uroporphyrin
and coproporphyrin by themethods of Schwartz,
Berg, Bossenmaier & Dinsmore (1960).
Protein content was determined by the method
of Lowry, Rosebrough, Farr & Randall (1951),
bovine serum albumin being used as standard.
Student’s t-test (two-sided) was used for
statistical analysis of significance, and possible
correlations were tested assuming two-variable
normal distribution (Dixon & Massey, 1957).
Results
Characterizationof subjects studied
Table 1 summarizes clinical histological
features of subjects studied. The control group
included a spectrum of subjects, some with
normal tests of liver function and histology who
denied ethanol ingestion, others with various
degrees of active hepatic inflammation, as
judged by clinical and bloodchemistry findings
and histology. Also included in the control
group were six men (subjects nos. 11-16) who
had drunk appreciable quantities of ethanol
[>2.2 mol (100g)/day] within 2 weeks of admission to hospital and who had last drunk
ethanol 1-17 days before biopsy. All six had
some degree of hepatic fatty change; some had
evidence of alcoholic hepatitis, but none had
cirrhosis.
The cirrhotic group comprised two patients
with post-necrotic cirrhosis of unknown cause
(subjects no. 17 and no. 18) and 10 chronic
alcoholics with Uennec’s cirrhosis of varying
severity and activity. The average quantity of
ethanol consumed daily [3.2 mol (147g)l within
TABLE
1. Summary of clinical and histological features of subjects
SGOT = serum glutamateoxaloacetate transaminase (Karmen units/l). The grading system 0-4+ is explained in
the Methods section.
Subject
group
Age
(years)
Liver
size* SGOT
(cm) (unitsll)
Serum alkaline
phosphatase
(i.u./l)
Control ( n = 16)
Mean
48.9
12
63
92
Range
27-67
8-22 13-231
33-307
Cirrhotic (n = 12)
Mean
50.7
15
78
149
Range
37-61
10-20 28-147
68-470
Non-cirrhotic, ingesting anticonvulsants (n = 5)
Mean
50.8
11
46
124
Range
42-62
10-14
19-91
83-167
Normal range
6-11
540
10-85
Serum
Serum
albumin globulin Fibrosis Necrosis
(do (0-4+) (W+)
Fat
Iron
(0-4+)
44
28-59
27
17-39
0.4
0-1
0.5
0-3
1.8
0-3
0.8
0-3
41
25-50
34
21-45
2.7
2-4
1.4
0-3
1.3
0-3
1.1
0-3
41
37-51
35-50
28
27-30
15-30
0.6
0-1
04
0-1
0
1.0
0-2
0
1.4
1-2
1-2
* Measured in right mid-clavicular line.
0
Human hepatic ALA synthase
513
TABLE
2. Serum iron concentrations and liver biochemistry
TlBC = serum total iron-binding capacity. Incubation was carried out without exogenous succinyl-CoAgenerating system or with exogenous succinyl-CoA-generating system.
Liver ALA synthase activity
(pmol of ALA h-' mg-' of protein)
Serum
Subjzct group
and no.
Control
1
2
3
4
5
6
7
8
9
10
11
12
13
14
I5
16
Mean
SD
Cirrhotic
17
18
19
20
21
22
23
24
25
26
27
28
Mean
SD
iron
(,umol/l)
Serum
TIBC
(pnolll)
6.9
107
8.2
33.6
465
37.5
40.7
64.2
69.6
65.3
-
7.8
24.4
10.0
16.4
25.5
-
12.7
-
-
57.5
57.6
42.4
69.8
67.6
-
6 25
-
-
18.4
127
57.7
12.0
26.4
43.6
26.4
61.8
6.4
209
17.3
300
65.6
55.1
47.3
38.4
19.1
19.1
207
7.3
-
-
-
Liver iron
(nmol/mg of protein)
Without
generating system
With generating
system
31.1
67
12.4
16,4
138
349
6.2
15.6
27-1
15.6
17.5
15.6
34.7
43.1
73.1
2.7
30.7
33.6
-
547
830
266
161
887
967
355
265
755
516
3 I9
38.5
5.1
42.4
28.4
5.5
181
143
-
179
23 1
I82
205
182
-
318
328
-
217
65
184
185
-
-
-
193
173
124
200
65.8
105
21.6
70,O
13.1
I10
-
-
55.1
5 41
10.2
18.2
41.6
37.5
29 1
193
49.9
-
Non-cirrhotic, ingesting anticonvulsants
29
15.5
43.3
30
16.4
60.5
31
32
13.3
622
33
27,3
58.7
18.1
56.2
Mean
SD
6.3
8.7
-
-
7-3
45.1
199
-
60.5
-
27.5
3 5I
22.9
240
220
-
P
Cirrhotic vs. control
Anticonvulsant vs.
control
-
634
731
620
373
548
252
1680
676
1721
566
779
269 1
-
1029
860
772
1800
1695
1297
664
792
858
1750
1140
1098
1128
379
0.02
018
0.06
062
<O W 5
014
0.18
065
0.8 1
< 0.005
514
H . L. Bonkowsky and Joanne S. Pomeroy
the 2 weeks preceding hospital admission by
those cirrhotics who had drunk during that
interval was closely similar to that consumed by
the drinking, non-cirrhotic control subjects
[3.1 mol (143 g)].
In addition, five non-cirrhotic subjects with
seizure disorders were studied. All had been ingesting diphenylhydantoin, 3 W 6 0 0 mg/day, for
4 months to 15 years before biopsy; two were
also ingesting phenobarbital (90 and 180 mg/
day) and one of these patients (subject no.
31) was ingesting mephenytoin as well (200
mg/day).
Overall, the data in Tables 1 and 2 are
interpreted as indicating that the control group
included subjects with hepatic functional abnormalities and serum and liver iron content
similar to those observed in the cirrhotic group
and the non-cirrhotic group ingesting anticonvulsants. Furthermore, a number of subjects
in the control group had drunk appreciable
amounts of ethanol shortly before admission
and biopsy and, except for lesser fibrosis and no
evidence of regenerative nodules, had histological findings similar to the patients with
alcoholic liver disease and Laennec's cirrhosis.
Validity of assay for ALA synthase
In preliminary experiments with homogenates
of mouse liver, the amount of ALA formed was
linear with respect to time of incubation for at
least 20 min and with respect to amount of liver
to at least 1.5 mg of protein/flask. This is in
accordance with the findings of Ebert et al.
(1970) and Bock, Krauss & Frohling (1971)
with the same method of assay for ALA synthase. The two human liver specimens obtained
by open wedge biopsy were sufficient for assays
at several homogenate concentrations, and these
results are summarized in Fig. 1. The amount of
ALA formed during a 20 min incubation, with
or without the exogenously added system for
generating succinyl-CoA, increased linearly as a
function of the amount of protein.
Fig. 1 also shows that addition of the system
for generating succinyl-CoA increased appreciably the amount of ALA formed, both by
normal and cirrhotic liver, although the enhancement was more pronounced in the latter.
As seen in Table 2, this phenomenon was observed consistently; the mean degree of enhancement for the six cirrhotic livers in which both
2400
0
c
/A
0.5
1.0
1-5
2.0
Protein (mg)
FIG.1. Formation of 8-aminolaevulinic acid as a function of amount of liver homogenate protein. Incubations were carried out aerobically at 37" C for 20 min in
25 ml Erlenmeyer flasks on a Dubnoff metabolic shaker.
0 , Homogenate of normal liver (subject no 6 ) without
exogenous system for succinyl-CoA generation; 0,
homogenate of the same normal liver, with exogenous
system for succinyl-CoA generation; A, homogenate of
cirrhotic liver (subject no. 28). without exogenous
system for succinyl-CoA generation; A , homogenate of
the same cirrhotic liver, with exogenous system for
succinyl-CoA generation. Vertical lines span the ranges
for two or more determinations.
assays were done was 7.5 (SD 3.9), compared
with eight control subjects in which it was 3.1
(SD 1.9), P < 0.05. Furthermore, this enhancing
effect of an exogenous succinyl-CoA-generating
system was not greater in the control or cirrhotic
groups who had drunk ethanol within 2 weeks
of hospital admission than in those who had not;
therefore the above difference between cirrhotic
and control subjects cannot be related simply to
the fact that a higher proportion of cirrhotic than
control subjects drank ethanol.
The activity of ALA synthase in the one
human liver we were able to study before and
after freezing at -85°C (from subject no. 6) was
relatively constant with or without a succinylCoA-generating system for at least 4 weeks.
Quick freezing-and-thawing of homogenate of
fresh human liver from this subject did not lead
to increased rates of formation of ALA under
our conditions of incubation. Freezing small bits
of human liver for at least 4 weeks and assaying
ALA synthase activity in whole-liver homogenates thus does not appear to produce spurious
results or results different from those obtained
Human hepatic ALA synthase
515
A
L
A
i
4
0 .
AA:.&
.
8
g
A
Time biopsy frozen before assay (doys)
Time since lost ingestion of ethanol(days)
.,
FIG.2. Scatter diagrams showing that activity of ALA synthase was not correlated with the time the
biopsy was frozen before assay (a) or with ethanol ingestion or time since last ethanol ingestion by
subjects before biopsy (b). Symbols are as in Fig. 1. with additional symbols: homogenate of livers
from subjects ingesting anticonvulsants, no exogenous system for succinyl-CoA generation; 0 , homogenate of livers from these subjects, with an exogenous system for succinyl-CoA generation. 'No EtOH'
indicates subjects who ingested ethanol rarely or not at all.
with homogenate of fresh liver. This conclusion
is further supported by the fact that, for all our
biopsies, there was no systematicchange in ALA
synthase activity as a function of the duration
the biopsy was frozen before assay (Fig. 2a).
Cirrhotic subjects
As shown in Table 2, in the absence of exogenously added succinic thiokinase and other
reagents for optimum generation of succinylCoA the mean rate of formation of ALA was
virtually identical in homogenates of control
and cirrhotic livers. However, when these reagents were added, the mean rate of ALA formation by homogenates of cirrhotic livers (1297
pmol of ALA h-' mg-' of protein, SD 664) was
significantly greater than the mean control value
(548, SD 252; P c 0.005). As shown in Fig. 2b,
there was no dependency of activity of ALA
synthase upon recent or remote ethanol ingestion. Furthermore, since control and cirrhotic
groups included subjects with similar degrees of
fatty infiltration and liver cell necrosis (Table
I), the difference in activity of ALA synthase
cannot be related to these features either, a
conclusion supported by the lack of correlation
between them and ALA synthase activity (data
not shown).
As shown in Fig. 3, activity of ALA synthase
was not closely correlated with hepatic iron
content either. Among 15 control subjects, the
correlation coefficient between maximal activity
of ALA synthase and liver iron content was
r = 0.53 (P = OM), whereas for 11 cirrhotic
subjects, r = 0.41 (P = 0.24). Thus it is not
surprising that the striking difference between
maximum observed activity of ALA synthase in
homogenates from cirrhotic compared with
control subjects was not related to a similar
difference in hepatic iron content (Table 2).
As seen in Table 3, despite the increased
activity of ALA synthase in the cirrhotic livers,
the mean urinary excretion rate of ALA was
lower in the cirrhotic subjects than in the control
group. The results in Table 3 also indicate that
any increased amounts of ALA which may have
been formed in cirrhotic subjects were not excreted as porphobilinogen. However, the cirrhotic subjects did excrete more uroporphyrin
H . L. Bonkowsky and Joanne S. Pomeroy
516
of porphobilinogen was also greater, but the
difference is not interpreted as being significant
(P = 0.12). Mean urinary porphyrin excretions
by subjects taking anticonvulsants were less than
those of control subjects, but the differences
were slight and not significant statistically.
Discussion
0
A
0
A
0
x
I
v)
4
cl
O
I
50
I
I
100
I
I
I50
Concn. of liver iron (nrnol of Fe/rng of protein)
FIG.3. Scatter diagram of activity of hepatic ALA synthase, in the presence of an exogenous succinyl-CoAgenerating system, vs. hepatic iron concentration. 0,
Control subjects; A , cirrhotic subjects; 0,subjects ingesting anticonvulsants For control subjects, the
regression equation is y = 3.87x+426, r = 0.53,
P = 004; for cirrhotic subjects the equation is
.v = 7.27x+930, r = 0.41, P = 0.24; for subjects ingesting anticonvulsants, the equation is y = 9.05x+ 840,
r = 0.55, P = 0.45 (Pis large since the sample size was
only four).
(P = 0.05) and coproporphyrin (P = 0.07)
than the control subjects, although the latter
difference was significant only at the 77; level.
Subjects ingesting anticonoulsants
As shown in Table 2, in two subjects who had
chronically ingested anticonvulsants, the activity
of hepatic ALA synthase, in the absence of a
exogenous system for generating succinyl-CoA,
was virtually identical with that in control
subjects. However, as for homogenates from
cirrhotic subjects, activities in homogenates from
anticonvulsant-treated subjects were significantly
higher than those of control subjects, when
assayed in the presence of the exogenous
succinyl-CoA-generating system. Furthermore,
as was also the case for the cirrhotic subjects,
there was no correlation between ALA synthase
activity and the degree of hepatic cell necrosis or
fat or iron deposition.
In contrast to subjects with cirrhosis, however,
those who chronically had ingested anticonvulsants excreted more ALA in the urine than did
control subjects (Table 3); their mean excretion
Our observations demonstrate the importance
of adding an exogenous system for generating
succinyl-CoA in order to assay maximal activity
of human ALA synthase in vitro. A similar
demonstration has been provided recently for
chick liver ALA synthase (Yoda el al., 1975).
In earlier studies of this activity, largely
carried out with rat liver, this was thought not to
be required (Marver et al., 1966), which led to
the use of assays with liver from other species,
including man, in which no exogenous system
for generation of succinyl-CoA was added.
However, Beattie and her associates (Patton &
Beattie, 1973; Beattie, Patton & Rubin, 1973)
have suggested that permeability of rat liver
mitochondria to succinyl-CoA is important in
modulating activity of ALA synthase, in
apparent disagreement with earlier findings
(Tschudy et al., 1964).
Although it is uncertain whether an exogenous
system for succinyl-CoA generation is required
only under conditions of low human liver
homogenate concentrations in oitro, as was the
case in our experiments, enhancement of ALA
synthesis by such a system raises a question
of whether activity of ALA synthase in oivo
can also be modulated by the supply of succinyl-CoA at sites of ALA synthase activity.
Our studies were not designed to address this
question, but they do provide the following
evidence that the increased activities of ALA
synthase we observed upon addition of the
succinyl-CoA-generating system relate to the
presence of more active enzyme rather than
simply increased substrate availability.
If the differences we observed between control
subjects and other groups related simply to
differences in supply of endogenous succinylCoA, one would expect these differences to disappear upon incubation with exogenous succinyl-CoA. In fact, however, quite the opposite
occurred. Furthermore, since freeze-thawing
homogenate from fresh control liver before incubation had no effect on ALA formation, the
Human hepatic ALA synthase
517
TABLE
3. Urinary excretion of porphyrins and their precursors
PBG = porphobilinogen; Uro = uroporphyrin; Copro = coproporphyrin.
Excretion (pmol/mol of creatinine)
Subject group and no.
Control
1
2
3
4
5
6
7
8
9
10
11
12
13
14
I5
16
Mean
SD
ALA
PBG
Uro
Copro
912
1190
1090
1840
583
282
305
120
277
106
397
286
1.09
1.91
1.36
046
6.44
2.36
4.02
2.10
2.52
2.16
18.5
4.67
4.15
31.3
3.49
19.4
19.0
10.2
5.04
47.5
14.5
22.5
49.8
30.9
21.1
14.2
1510
1350
444
1030
924
1240
1130
1450
1090
1520
97 1
1170
1190
299
144
218
314
166
244
156
27 1
191
385
25 1
102
Cirrhotic
17
1160
18
19
20
21
22
23
24
25
26
27
28
Mean
1110
SD
185
236
538
51
161
253
198
189
217
26 1
261
232
118
743
1260
634
927
648
1110
1040
1210
969
670
957
230
Non-cirrhotic. ingesting anticonvulsants
29
30
1790
31
1390
32
33
1630
Mean
SD
2.86
4.36
2.59
9.12
3.13
2.3 I
7.90
8.17
9.67
3.81
5.00
4.56
5.29
2.69
15.1
30.1
I68
15.4
802
19.0
11.1
22.5
51.8
56.0
23.3
39.6
24.5
325
206
-
-
-
454
315
3.68
2.44
16.2
12.3
255
341
102
2.04
2.72
086
19.9
-
1600
20 1
8.44
3.68
7.76
3.42
3.48
2.34
-
247
-
-
161
3.80
P
Control vs. cirrhotic
Control vs. anticonvulsant
lesser activities of ALA synthase in control livers
cannot be ascribed to diminished membrane
permeability of endogenous or exogenous
succinyl-CoA. The simplest explanation for our
findings appears to be that less active enzyme
005
0.03
0.94
012
005
067
007
025
was present in homogenates from control subjects so that, under optimum conditions of
assay, less ALA was produced by these homogenates.
In light of the above considerations, we believe
518
H. L. Bonkowsky and Joanne S. Pomeroy
activities of hepatic ALA synthase in vivo are
increased in cirrhotic subjects and in subjects
ingesting anticonvulsants. This conclusion is
consonant with the known effects of anticonvulsants upon activity of the enzyme in experimental systems (Tschudy, 1974; Granick, 1966)
as well as the excess porphyrinuria or overt
porphyria which occurs in cirrhotic subjects
with increased frequency (Doss et al., 1972).
Several findings from our studies should be
borne in mind in considering this conclusion. 1.
Use of a control group which included subjects
who drank ethanol in significant amount and
subjects with abnormalities of liver-function tests
and degrees of hepatic iron deposition similar to
the cirrhotic subjects excludes these factors as
satisfactory explanations for the differences
observed. Furthermore, save for the absence of a
cirrhotic pattern, control subjects had histological findings similar to those for cirrhotic
subjects. The control group used seems appropriate to assess effects of cirrhosis or anticonvulsant ingestionper se upon the variables studied. 2 .
The increase in activity of ALA synthaselmg of
protein in cirrhotic livers or livers of anticonvulsant-treated patients was not a compensatory
phenomenon, as might have occurred if the total
hepatic mass were decreased in these subjects.
Table 1 shows that mean liver size of the cirrhotic subjects was increased, although it was not
different from control values among those
chronically treated with anticonvulsants. N o
subject studied had an abnormally small liver
and no inverse correlation between estimated
liver size and measured ALA synthase activity
was observed. Thus our data reflect an increase
in total capacity for hepatic ALA synthesis in
experimental subjects compared with control
subjects. 3. For reasons already given, the
difference in activity of ALA synthase between
control and the other subjects does not appear
to result from differential permeation of succinylCoA to the active site.
The cause of the increase in ALA synthase
activity in cirrhotic liver is unknown. A previous
suggestion was that it depended upon alterations
in sex steroid concentrations (Levere, 1967),
which are common in cirrhosis (Chopra,
Tulchinsky & Greenway, 1973;Kley, Nieschlog,
Wiegelmann, Solbach & Kriiskemper, 1975).
Although a wide variety of sex steroids are
capable of inducing mammalian hepatic ALA
synthase (Edwards & Elliott, 1975), this ex-
planation seems unlikely since the effect
observed is relatively non-specific, being produced by a large number of C-19 and C-21 sex
steroids (Edwards & Eliott, 1975), some of
which are increased in concentration and some
decreased in plasma of cirrhotic patients
(Chopra el al., 1973; Kley et al., 1975). Furthermore, these steroids increase activity of mammalian hepatic ALA synthase activity only at
concentrations about four orders of magnitude
higher than those which occur in plasma from
normal or cirrhotic patients (Kley et al., 1975),
making it improbable that this effect is of physiological importance.
In contrast, di- and tri-hydroxycoprostane,
intermediates in the synthesis of bile acids, are
more likely to be endogenous physiological inducers of hepatic ALA synthase since their
effects are relatively specific and occur at concentrations which could reasonably occur in
ciuo (Javitt, Rifkind & Kappas, 1973). In
cirrhosis, the rates of synthesis of chenodeoxycholic acid and particularly of cholic acid are
decreased (Vlahcevic, Buhac, Farrah, Bell &
Swell, 1971; Vlahcevic, Juttijudata, Bell &
Swell, 1972). Although the cause of this is unknown currently, it may be due to defective
side-chain oxidation of the di- and tri-hydroxycoprostane, leading to their accumulation in the
liver in concentrations sufficient to induce ALA
synthase. The increased activity of ALA synthase in cirrhotic liver could be a secondary
metabolic aberration.
A central role has been postulated for cytochrome P-450 in control of activity of hepatic
ALA synthase and haem biosynthesis (Tschudy,
1974). For example, many chemicals which induce ALA synthase may do so by virtue of a
primary effect on cytochrome P-450, leading to
relative lack of ‘free haem’, the prosthetic group
of P-450, and secondarily to induction of ALA
synthase. This concept provides a parsimonious
explanation for the increased activity of ALA
synthase we observed in homogenates from
subjects pretreated with hydantoins (and barbiturates), compared with control subjects,
since these drugs are well-known inducers of
cytochrome P-450 and have been shown to induce hepatic ALA synthase in experimental
animals (Tschudy, 1974). Previously, strong
circumstantial evidence from patients with
hereditary hepatic acute prophyrias (Tschudy,
1974) had incriminated such inducers of cyto-
H u m n hepatic ALA synthase
chrome P-450 as inducers of human ALA synthase as well. Our observations indicate that
increased activity of hepatic ALA synthase does
occur in human liver and can be detected in liver
from subjects treated with these drugs who, unlike porphyric patients, do not have defects in
activity of a distal enzyme of the haem synthetic
pathway. Thus, in non-porphyric man as in
other mammals, inducers of cytochrome P-450
can also increase ALA synthase activity.
In contrast, in homogenates of cirrhotic livers
levels of microsomal cytochrome P-450 and
mixed-function oxidase activities are decreased
(Schoene, Fleischmann & Remmer, 1972). Our
findings of increased ALA synthase activity in
homogenates from cirrhotic livers may thus
seem paradoxical. However, the level of cytochrome P-450 is a balance between the rate of
formation and the rate of degradation, and were
the latter to be increased in cirrhosis, as occurs
in other circumstances (Meyer & Marver, 1971 ;
Marver & Schmid, 1972), coexistence of decreased levels of cytochrome P-450 and increased activity of ALA synthase (and haem
synthesis)would readily be understood. Furthermore, whereas normally utilization of ALA for
eventual synthesis of haem is highly efficient,
this may not be so in diseased liver. If there were
a decrease in the efficiency with which ALA or
other haem precursors are converted into haem,
increased rates of synthesis of ALA would be
required to maintain any given steady state of
hepatic haem production. The increased urinary
excretion of porphyrins, repeatedly observed in
cirrhotic patients (Tschudy, 1974; Doss et al.,
1972) and confirmed here (Table 4), may be one
reflection of such decreased efficiency of haem
precursor utilization, and our and other excretion data suggest a partial block at the level
of uroporphyrinogen utilization, perhaps with
an increase in porphobilinogen utilization, a
combination previously observed in other circumstances (Kushner et al., 1972, 1975).
However, the increased porphyrin excretions
commonly seen in cirrhotic subjects are relatively small and do not approach the amounts
expected if all the additional capacity for ALA
formation were excreted as excess urinary
porphyrinogen or porphyrin. [For example,
assuming a liver wet weight of 1500 g, 10% of
which is protein, from our data one would
calculate that in control subjects capacity for
hepatic ALA synthesis is about 2 mmol/day (55
519
mmol h-' g-' wet w t . x l 5 0 0 g x 2 4 h). In
cirrhotic subjects, the mean value is 2.4 times
this, or 4.8 mmol/day. If all the excess of 2.8
mmol of ALA were excreted as porphyrins, the
degree of porphyrinuria would be enormous (cf.
Table 3).]
Thus, although there may be some relation in
cirrhotic subjects between increased activity of
hepatic ALA synthase and porphyrinuria, the
latter is not sufficient quantitatively to explain
the former. Some other avenue for disposition
of the increased amounts of ALA synthesized
would be required, and it is tempting to speculate that the rate of hepatic haem degradation is
increased and perhaps is reflected in decreased
hepatic cytochrome P-450 levels as outlined
above. An alternative possibility, for which,
however, there is no evidence currently, could
involve increased activity of porphobilinogen
oxygenase, which has been observed in rat livers
under conditions associated with increased
activities of ALA synthase (Tomaro, Frydman
& Frydman, 1973).
One further possible explanationfor increased
activity of ALA synthase in cirrhotic liver
deserves mention: the increase may reflect hepatocyte regenerative activity,which is more active
in cirrhotic liver than in normal liver. Regenerating liver develops many metabolic attributes
similar to foetal liver, and this may be true for
ALA synthase activity, which has a higher
activity (Woods, 1974) and different regulatory
characteristics(Woods & Murthy, 1975) in foetal
liver than in liver from normal adults.
The increase in activity of ALA synthase we
observed in homogenates of cirrhotic liver
(mean activity = 2.4 x control value) is less than
the 5-15-fold increases previously found by
Levere (1967) in mitochondria isolated from five
of seven cirrhotic livers. Possible reasons for this
difference include the following: 1, his subjects,
shortly before biopsy, received a barbiturate,
which may have been metabolized more slowly
in the cirrhotic than in the control subjects
(Held, von Olderschauen & Remmer, 1970) and
therefore may have induced ALA synthase
acutely to higher activity; 2, if the findings of
Patton & Beattie (1973) are confirmed, the
possibility will arise that mitochondria as isolated by Levere from livers were more permeable
to citrate than those from control subjects, or
the endogenous generation of succinyl-CoA
from citrate was greater, leading to increased
520
H . L. Bonkowsky arid Joanne S. Pomeroy
activity of the enzyme in uitro, perhaps without
an equivalent increased amount of enzyme or
activity in vivo. For reasons already given, the
increased activities we observed cannot be. explained on these bases.
Acknowledgments
We thank Dr John Gilbert and Dr Jane
Brisbane for histological review, Dr June
Lascelles for supplying us with a slant culture of
H-5strain of R. spheroides, and Dr Nicholas
Jacobs, in whose laboratory this strain has been
maintained and extracts prepared. This work
was carried out in part during the tenure of a
USVA Research Associateship, awarded to
H.L.B. and supported by VA funds, MRIS 1023
(01).
References
BARRY,
M. & SHERLOCK,
S.(1971) Measurement ofliver
iron concentration in needle-biopsy specimens.
Lancet, i, 100-103.
BEATTIE, D.S., PATTON,G.M. & RUBIN,E. (1973) The
control of 8-aminolevulinic acid synthetase in rat
liver mitochondria. Effect of pyrazole on the apparent
'induction.' Enzyme, 16, 252-257.
BOCK, K.W., KRAUSS,E. & FRBHLING,
W. (1971)
Regulation of 8-aminolevulinic acid synthetase by
drugs and steroids in oioo and in isolated perfused rat
liver. European Journal of Biochemistry, 23, 366-371.
CHOPRA,I.J., TULCHINSKY,
D. & GREENWAY,
F.L.
(1973) Estrogen-androgen imbalance in hepatic
cirrhosis. Studies in 13 male patients. Annals of
Internal Medicine, 19, 198-203.
CONN,H.O. (1972) Portacaval anastomosis and hepatic
hemosiderin deposition: A prospective controlled
investigation. Gastroenterology, 62, 61-72.
CONN,H.O. (1975) Cirrhosis, In: Diseases of the Liver.
4th edn, chapter 26. Ed. Schiff, L. Lippincott.
Philadelphia.
F.J., JR (1957) Introduction to
DIXON,W.J. & MASSEY,
Statistical Analysis, 2nd edn. McGraw-Hill, New
York.
Doss, M., LOOK, H., HENNING,P., NAWROCKI,
P.,
SCHMIDT,
A., DBLLE,W., KORB,G., LUDERS.C.J. &
STROHMEYER,
G. (1972) Hepatic porphyrins and
urinary porphyrins and porphyrin precursors in liver
cirrhosis. Klinische Wochenschrift, 50, 1025-1032.
EBERT,P.S., TSCHUDY,
D.P., CHOWDHRY,
J.N. &
CHIRIGOS,
M.A. (1970) A simple micro method for the
direct determination of 8-amin0[~~C]levulinic
acid
production in murine spleen and liver homogenates.
Biochimica et Biophysica Acta, 208, 236-250.
EDWARDS,
A.M. & ELLIOTT,
W.H. (1975) Induction of
8-aminolevulinic acid synthetase in isolated rat liver
cells by steroids. Journal of Biological Chemistry, 250,
2750-2755.
GRANICK,
S. (1966) The induction in oitro of the synthesis of 8-aminolevulinic acid synthetase in chemical
porphyria: a response to certain drugs, sex hormones,
and foreign chemicals.Journalof Biological Chemistry,
241, 1359-1375.
HELD,H., VON OLDERSCHAUEN,
H.F. & REMMER,
H.
(1970) Der Abbau von Pentobarbital bei Leberschaden. Klinische Wochenschrift, 48,565-567.
JAVIIT, N.B.. RIFKIND,A. & KAPPAS,A. (1973)
Porphyrin-heme pathway: regulation by intermediates in bile acid synthesis. Science, 182, 841-842.
S., GILVARG.
C., CORI,0. & OCHOA,S.
KAUFMAN,
(1953) Enzymatic oxidation of a-ketoglutarate and
coupled phosphorylation. Journal of Biological
Chemistry, 203,869-888.
KLEY, H.G.. NIESCHLOG.E., WIEGELMANN,
W.,
SOLBACH,
H.G. & KRUSKEMPER,
H.L. (1975) Steroid
hormones and their binding in plasma of male
patients with fatty liver, chronic hepatitis, and liver
cirrhosis. Acta Endocrinologica, 79, 275-285.
KUSHNER,
J.P., LEE,G.R. & NACHT,S. (1972) The role
of iron in the pathogenesis of porphyria cutanea tarda.
An in oitro model. Journal of Clinical Inoestigation, 51,
3044-305 I.
KUSHNER,
J.P., STEINMULLER,
D.P. & LEE,G.R. (1975)
The role of iron in the pathogenesis of porphyria
cutanea tarda. 11. Inhibition of uroporphyrinogen
decarboxylase. Journal of Clinical Investigation, 56,
66 1-667.
LEVERE.R.D. (1967) Porphyrin synthesis in hepatic
cirrhosis: increase in 8-aminolevulinic acid synthetase
Biochemical Medicine. 1, 92-99.
LOWRY,O.H., ROSEBROUGH,
N.J., FARR,A.L. &
RANDALL,
R.J. (1951) Protein measurement with the
Folin phenol reagent. Journal of Biological Chemistry,
193.265-275.
MARVER,
H.S. & SCHMID,
R. (1972) The porphyrias. In:
The Metabolic Basis of Inherited Disease, 3rd edn,
chapter 45. Ed. Stanbury, J.B., Wyngaarden, J.B. &
Fredrickson, D.S. McGraw-Hill, New York.
MARVER.
H.S., TSCHUDY,D.P., PERLROTH,
M.D. &
COLLINS,
A. (1966) 8-Aminolevulinic acid synthetase
1. Studies in liver homogenates. Journal of Biological
Chemistry, 241, 2803-2809.
MAUZERALL,
D. & GRANICK.
S. (1956) The occurrence
and determination of 8-aminolevulinic acid and
porphobilinogen in urine. Journal of Biological
Chemistry, 219, 435446.
MEYER,U.A. & MARVER,H.S. (1971) Chemically
induced porphyria: increased microsomal heme turnover after treatment with allylisopropylacetamide.
Science, 171, 64-66.
PATTON,
G.M. & BEAITIE,D.S.(1973) Studieson hepatic
8-aminolevulinic acid synthetase. Journal of Biological
Chemistry, 248, 44674474.
PIPER,W.N., CONDIE,
L.W. & TEPHLY,
T.R. (1973) The
role of substrates for glycine acyltransferase in the
reversal of chemically induced porphyria in the rat.
Archives of Biochemistry and Biophysics, 159. 671677.
RUBM,E.,BACCHIN,
P.,GANG,H. & LIEBER,
C.S. (1970)
Induction and inhibition of hepatic microsomal and
mitochondria1 enzymes by ethanol. Luboratory
Investigation, 22, 569-580.
SCHEUER,
P.J., WILLIAMS,
R. & MUIR, A.R. (1962)
Hepatic pathology in relatives of patients with
haemochromatosis. Journal of Pathology and Bacteriology, 84, 53-64.
SOHOENE,
B.. FLEISCHMANN.
R.A. C REMMER,
H. (1972)
Determination of drug metabolizing enzymes in
needle biopsies of human liver. European Journal of
Clinical Pharmacology, 4.65-73.
SCHWARTZ,
S., BERG,M.H..BOSSENMAIER,
I. & DINSMORE, H. (1960) Determinations of porphyrins in
biological materials. Methods in Biochemical Analysis,
8, 221-293.
Human hepatic ALA synthase
SHANLEY,
B.C., ZAIL, S.S. & JOUBERT,S.M. (1968)
Effect of ethanol on liver &aminolaevulinate synthetase in rats. Loncet, I, 70-71.
ST~IN,J.A., TSCHUDY,
D.P., CORCORAN,
L.P. &
COLLINS,
A. (1970) 6-Aminolevulinic acid synthetase.
111. Synergistic effect of chelated iron on induction.
Journal of Biological Chemistry, 245, 2213-221 8.
TOMARO,
M.L., FRYDMAN,
R.B. & FRYDMAN,
B. (1973)
Porphobilinogen oxygenase from rat liver: induction,
isolation, and properties. Biochemistry, 12, 52635268.
TSCHUDY,
D.P. (1974) Porphyrin metabolism and the
porphyrias. In: Duncan’s Diseases of Metabolism, 7th
edn, chapter 12. Ed. Bondy, P.K.& Rosenberg, L.E.
Saunders, Philadelphia.
TSCHUDY,D.P., WKLLAND,F.H., COLLINS,A. &
HUNTER,G., JR (1964) The effect of carbohydrate
feeding on the induction of 6-aminolevulinic acid
synthetase. Metabolism, 13, 396-406.
VLAHCEVIC,
Z.R., BUHAC,I., FARRAH,
J.T., BELL,C.C.,
521
JR & SWELL,L. (1971) Bile acid metabolism in
patients with cirrhosis. 1. Kinetic aspects of cholic
acid metabolism. Gastroenterology, 60,491-498.
VLAHCKVIC,A.R., JUTTIJUDATA,
P.. BELL,C.C.. JR &
SWELL,L. (1972) Bile acid metabolism in patients
with cirrhosis. 11. Cholic and chenodeoxycholic acid
metabolism. Gastroenterology, 62, 1174-1 181.
WOODS,J.S. (1974) Studies on the role of heme in the
regulation of &aminolevulinic acid svnthetase during
fecal hepatic development. Molecul& Pharmacologi
10. 389-397.
WOODS.J.S. & MURTHY,
V.V. (1975) 6-Aminolevulinic
acid synthetase from fetal rat liver: studies on the
partially purified enzyme. Molecular Pharmacology,
11, 70-78.
YODA,B., SCHACHTER,
B.A. & Isru~u,L.G. (1975)
Delta-aminolevulinic acid synthetase assay in chickrn
liver homogenates and particulate fractions. Analytical
Biochemisrry. 66. 221-233.