Departments of Medicine (R.W.L., O.S.G., J.A.P., K.D.T., H.L.B.)
and Biochemistry and Molecular Biology (H.L.B.) of the University of
Massachusetts Medical School; and The Center for Study of Disorders of
Iron and Porphyrin Metabolism of UMass Memorial Health Care (R.W.L.,
O.S.G., J.A.P., K.D.T., A.W., H.L.B), Worcester, Massachusetts
Patients with any of the acute porphyrias may suffer from acute
attacks. If these patients are treated with certain drugs, such as
barbiturates, the likelihood of developing an attack is increased.
Patients treated with antidepressants or benzodiazepine-type anxiolytics also could be placed at increased risk of developing porphyric attacks because little is known about the potential for some
of these drugs to induce attacks. Primary cultures of chick embryo
liver cells were used to study the effects of selected antidepressants
and anxiolytics on porphyrin accumulation. Cells were treated with
desferrioxamine (to partially block heme synthesis, simulating
conditions encountered in porphyric patients) and increasing concentrations (3.16-1000 µM) of the evaluated drugs. Twenty hours later, porphyrin accumulation was measured. The drugs included four
antidepressants and five benzodiazepine-type anxiolytics. The
antidepressants bupropion and nefazodone significantly increased porphyrin accumulation when given with desferrioxamine, whereas neither
fluoxetine nor paroxetine increased porphyrin accumulation. The
benzodiazepine-type anxiolytic agents oxazepam, lorazepam, diazepam,
triazolam, and midazolam all significantly increased porphyrin
accumulation when given with desferrioxamine. Dose-response studies
showed that diazepam, midazolam, and triazolam produced significant
increases even at the lowest concentration tested (3.16 µM), whereas
lorazepam and oxazepam required higher concentrations (
10 µM).
These studies suggest that patients with acute porphyrias may be at
greater risk for developing porphyric attacks when treated with
bupropion or nefazodone compared with fluoxetine or paroxetine, and
that the evaluated benzodiazepine derivatives should be administered with caution. Among the latter, low doses of lorazepam and oxazepam may
be safer than those of diazepam, midazolam, and triazolam.
 |
Introduction |
The
biosynthesis of heme and its precursors is normally tightly regulated,
and this process typically results in very little accumulation, or
excretion, of either intermediates or side-products from this pathway
(Kappas et al., 1995
; Hahn and Bonkovsky, 1998). In most organs, this
pathway is primarily regulated at the level of the first
heme-biosynthetic enzyme, 5-aminolevulinate (ALA) synthase (EC
2.3.1.37). This enzyme has been studied extensively in liver cells
where its level is controlled by a small, but critical, heme pool that
has been called the "regulatory heme pool" (Hahn and Bonkovsky,
1998). A decrease in the amount of heme in this pool produces an
up-regulation of ALA synthase, whereas a sufficiency, or excess, of
heme in the pool exerts the opposite effect (Kappas et al., 1995; Hahn
and Bonkovsky, 1998). Heme represses ALA synthase by decreasing the
stability of the ALA synthase mRNA (Drew and Ades, 1989; Hamilton et
al., 1991; Cable et al., 1994b; 1996) and by decreasing the rate of
import of ALA synthase into mitochondria (Ades and Harpe, 1981; Lathrop
and Timko, 1993).
The porphyrias are a group of related disorders whose underlying cause
is an acquired or hereditary defect in the activity of one of the
enzymes of heme biosynthesis distal to ALA synthase. In the acute
porphyrias (ALA dehydratase deficiency porphyria, acute intermittent
porphyria, hereditary coproporphyria, variegate porphyria), patients
may develop acute porphyric attacks. These are characterized typically
by abdominal pain, obstipation, nausea, vomiting, and other
neurovisceral manifestations (Hahn and Bonkovsky, 1998). During such
attacks, there is marked induction of ALA synthase. This may be due to
a deficiency of heme in the regulatory heme pool, and by either
increased breakdown of heme in hepatocytes or increased demand for
heme, particularly for formation of new molecules of cytochrome P-450
and possibly other hemoproteins. Various drugs and chemicals are among
the major effectors of P-450 induction and/or heme depletion within
hepatocytes, and such drugs and chemicals continue to be major causes
of acute attacks of porphyria (Moore, 1980; Hahn and Bonkovsky, 1998).
Some porphyrogenic drugs and chemicals induce ALA synthase by a
mechanism that is not solely dependent upon changes in the regulatory
heme pool (Hamilton et al., 1988; Ryan and Ades, 1989). Regardless of
the precise proximate cause for induction of ALA synthase, such
induction is a sine qua non for acute porphyric attacks.
A continuing question in therapeutics for patients with acute porphyria
is "Which drugs are `safe' for use in these diseases?" This
question is of particular importance with respect to choice of
analgesics, antihypertensives, anxiolytics, antidepressants, and
anticonvulsants because patients with acute porphyria suffer from acute
and chronic pain syndromes, systemic arterial hypertension, neuropsychiatric disorders, and seizures more frequently than nonporphyric patients. Thus, appropriate therapy for these
complications is often required. In this article, we report effects of
selected antidepressants and anxiolytics on porphyrin accumulation in a relevant and robust experimental model of acute porphyria.
| |
Experimental Procedures |
Materials.
The reagents and supplies used for preparing and
maintaining primary cultures of chick embryo liver cells (CELCs) were
as described in Hahn et al. (1997). The drugs tested were purchased from the following sources: bupropion and diazepam (Research
Biochemicals Inc., Natick, MA); nefazodone (100-mg tablets of Serzone;
Bristol-Myers Squibb, Co., Princeton, NJ); fluoxetine (20-mg tablets of
Prozac; Dista Products Company, Indianapolis, IN); paroxetine (40-mg
tablets of Paxil; SmithKline Beecham Pharmaceuticals, Philadelphia,
PA); oxazepam (10-mg USP capsules; Geneva Pharmaceuticals, Inc.,
Broomfield, CO); lorazepam (Ativan injection; Wyeth Laboratories,
Philadelphia, PA); triazolam (0.25-mg USP tablets; Roxane Laboratories,
Inc., Columbus, OH); and midazolam (Versed injection; Roche
Laboratories, Nutley, NJ).
Cell Cultures.
Primary CELC cultures were prepared and
maintained as described in Hahn et al. (1997). Five hours after the
first change of the culture medium, cells were treated with selected
concentrations of drugs and 250 µM desferrioxamine (DES), and this
treatment continued for 20 h. Cultures were incubated in the dark
at 37°C under an atmosphere of 5% (v/v) CO2 in air.
Sodium phenobarbital, midazolam, lorazepam, and diazepam were dissolved
in sterile water before their addition to the culture. The rest of the
drugs that were evaluated were dissolved in dimethyl sulfoxide (DMSO).
The volume of DMSO added to the cultures never exceeded 1 µl/ml
culture medium. In the culture system used, DMSO added in this volume does not effect porphyrin synthesis or accumulation (Bonkovsky et al.,
1992; Cable et al., 1994a).
Assay of Porphyrins.
To determine the amount of porphyrin
accumulation, cells and medium were harvested together, and porphyrins
were extracted and assayed as described previously (Hahn et al., 1997)
with a spectrofluorometric procedure (Grandchamp et al., 1980). Results obtained with this method have repeatedly and consistently been confirmed by studies with HPLC (Sinclair et al., 1986; Hahn et al.,
1997).
Assay of Proteins.
Protein concentrations were measured in
sonicates of cells plus medium with a Coomassie blue-based assay
(Bio-Rad, Hercules, CA), adapted to a microtiter plate technique. BSA
was used as the standard. The absorbance of the samples at 570 nm was
measured at room temperature in a Thermomax plate reader (Molecular
Devices Corp., Menlo Park, CA).
Statistical Procedures.
For each drug and each concentration
studied, triplicate samples were treated and assayed, and all studies
were performed in at least two independent experiments producing
similar results. Both positive (phenobarbital plus DES) and negative
(DMSO alone) controls were run in each experiment. The mean ± S.E. of total porphyrins that accumulated in the presence of DMSO alone
(1 µl/ml media) was 36 ± 1.2 ng/mg protein; in the presence of
DES alone (250 µM) was 196 ± 14.8 ng/mg protein; and in the presence
of phenobarbital (2 mM) plus DES (250 µM) was 1516 ± 59.4 ng/mg
protein, for the nine sets of data presented (n = 3 observations per data set for each of these treatments). Results of
typical experiments are presented in the figures with values of the
means + S.E., n = 3. Preliminary evaluation
revealed that the data were distributed normally. Thus, statistical
analyses were performed by ANOVA with the aid of SAS version 6.12 software (SAS Institute, Cary, NC). Pairwise comparisons were evaluated
for differences with the procedure of Tukey and Kramer.
P values <.05 were considered significant.
| |
Results |
Porphyrogenicity of Selected Antidepressants.
The abilities of
bupropion or nefazodone (administered alone or in combination with 250 µM DES) to increase porphyrin accumulation in the experimental model
system of porphyria are shown in Fig. 1.
Treatment with either of these antidepressants alone, at concentrations of 1000 µM, did not cause any significant increase in porphyrin accumulation compared with the porphyrin accumulation in cells treated
with the vehicle (DMSO). The treatment with the combination of
bupropion plus DES increased porphyrin accumulation at the three lower
concentrations tested (up to 31.6 µM; Fig. 1A). At concentrations
>31.6 µM, the combination of bupropion plus DES may have been toxic
to the cells, as indicated by a decrease in porphyrin accumulation. The
combination of DES plus nefazodone resulted in a dose-dependent
increase in porphyrin accumulation, up to 31.6 µM nefazodone, and
this porphyrin accumulation was nearly as large as that caused by the
positive control [DES plus phenobarbital (PB); Fig. 1B]. Higher
concentrations of nefazodone (>31.6 µM) resulted in an abrupt
decline in porphyrin accumulation that was due to cellular toxicity to
the CELCs. This abrupt change in porphyrin levels for concentrations of
nefazodone between 31.6 and 100 µM was observed in two independent
experiments. In all of these studies, the combination of DES plus PB
was included as a positive control, and this treatment always produced
significantly increased porphyrin accumulation compared with treatment
with DES alone.
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 1.
Effects of antidepressants bupropion (A) or
nefazodone (B) on porphyrin accumulation. CELCs were treated and
harvested, and porphyrins measured as described in Experimental
Procedures. The indicated concentrations of the test drugs and
DES (250 µM) were added to the cultures 20 h before harvest.
Data represent means + S.E., n = 3, except for the
cells treated with 31.6 µM nefazodone and 250 µM DES (mean + range;
n = 2). *, differs from DES only control,
p < .05.
|
|
The effects of treatment with fluoxetine and paroxetine, two selective
serotonin reuptake inhibitors, are shown in Fig.
2, A and B, respectively. The combination
of fluoxetine plus DES did not increase porphyrin accumulation compared
with treatment with DES alone. The combination of DES plus the higher
concentrations of fluoxetine (>10 µM) caused a decrease in porphyrin
accumulation, possibly due to toxicity. As shown in Fig. 2B, the
combination of DES plus paroxetine did not result in increased
porphyrin accumulations (compared with DES alone) at any of the
concentrations tested.
View larger version (24K):
[in this window]
[in a new window]
|
Fig. 2.
Lack of effect of antidepressants fluoxetine (A) or
paroxetine (B) on porphyrin accumulation. CELCs were treated and
harvested, and porphyrins measured as described in Experimental
Procedures. The indicated concentrations of the test drugs and
DES (250 µM) were added to the cultures 20 h before harvest.
Data represent means + S.E., n = 3. *, differs
from DES only control, p < .05.
|
|
Porphyrogenicity of Selected Benzodiazepine-Type Anxiolytic
Agents.
The porphyrogenic effects of oxazepam and lorazepam are
shown in Fig. 3, the porphyrogenic
effects of diazepam and triazolam are shown in Fig.
4, and the porphyrogenic effects of
midazolam are shown in Fig. 5. When given
in combination with DES, all five of these compounds significantly
increased porphyrin accumulation in CELCs for at least two of the
concentrations tested. Diazepam, midazolam, and triazolam produced
significant increases even at the lowest concentration tested (3.16 µM), whereas lorazepam and oxazepam required higher concentrations
(10 µM) to produce a significant effect. Also, at the higher
concentrations tested, all five of these compounds showed decreased
porphyrin accumulation, indicating that the combination of DES plus
these benzodiazepine-type compounds was toxic to the cells. Treatment
with 1000 µM lorazepam alone resulted in a small, but statistically
significant, increase in coproporphyrin accumulation compared with
treatment with vehicle (DMSO) alone.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 3.
Effects of oxazepam (A) or lorazepam (B) on porphyrin
accumulation. CELCs were treated and harvested, and porphyrins measured
as described in Experimental Procedures. The indicated
concentrations of the test drugs and DES (250 µM) were added to the
cultures 20 h before harvest. Data represent means + S.E.,
n = 3. *, differs from DES only control,
p < .05;  , differs from DMSO only
control, p < .05.
|
|
View larger version (29K):
[in this window]
[in a new window]
|
Fig. 4.
Effects of diazepam (A) or triazolam (B) on porphyrin
accumulation. CELCs were treated and harvested, and porphyrins measured
as described in Experimental Procedures. The indicated
concentrations of the test drugs and DES (250 µM) were added to the
cultures 20 h before harvest. Data represent means + S.E.,
n = 3. *, differs from DES only control,
p < .05.
|
|
View larger version (35K):
[in this window]
[in a new window]
|
Fig. 5.
Effects of midazolam on porphyrin accumulation. CELCs
were treated and harvested, and porphyrins measured as described in
Experimental Procedures. The indicated concentrations of
midazolam and DES (250 µM) were added to the cultures 20 h
before harvest. Data represent means + S.E., n = 3. *, differs from DES only control, p < .05.
|
|
| |
Discussion |
The purpose of the present study was to determine whether
treatment of CELCs with selected antidepressants and
benzodiazepine-type anxiolytic agents, administered alone or in
combination with DES, would increase porphyrin accumulation. This
information may be useful in guiding the selection of drugs prescribed
for patients with acute porphyria.
Predictions of whether a given drug will be safe for use with patients
with acute porphyria are often based on clinical experience with
porphyric patients, studies in laboratory animals, or studies in
cultured cells. Much of the published information from these sources
has been reviewed and assembled into lists of drugs that are
"safe," "not safe," or "contentious" (Eales, 1979; Moore, 1980; Moore and McColl, 1989; Dover et al., 1994; Ashley, 1996; Moore
and Hift, 1997; Gorchein, 1997; Kalman and Bonkovsky, 1998). Some of
this information is available on the World Wide Web (e.g., at
www.uct.ac.za/depts/liver/drugname.htm). Also, an international Committee on the Review of Porphyrinogenicity of Drugs has been established to facilitate the gathering and dissemination of this information.
The usefulness of these lists, of course, depends on the quality of the
information on which they are based. Although clinical experience with
porphyric patients is the best approach, it is hampered by several
factors. The number of subjects available for study at any given time
is almost always small, and the intentional exposure of these
individuals to medications of unknown potential to cause porphyric
attacks would be medically risky and ethically unsound. Often, when
patients with porphyria are hospitalized for reasons unrelated to their
porphyria (e.g., childbirth, surgery, etc.), they are treated with
combinations of drugs, making it difficult to determine which (if any)
of the drugs used was the actual causative agent in the event of a
resulting acute attack. Also, the suspect drug may not even have been
the cause of the attack because there are many non-drug-precipitating
factors, including infection, fasting, stress (from surgery or other
causes), and changes in hormonal balance.
The use of intact animals to study the porphyrogenic effects of drugs
also is limited by a number of factors. These include the expense and
difficulty of conducing such studies, particularly if a range of doses
for many drugs is to be tested. The means by which the animals are made
partially deficient in heme synthesis also could affect the results. In
the past, this has been done with chemical agents (Goerz et al., 1997),
but as mice with inherited defects in heme biosynthetic enzymes
(Lindberg et al., 1996) become available for such studies, they might
well become the test system of choice. The problem of interspecies
variation in response to many drugs remains, making valid comparisons
between human and animal systems difficult.
We chose to investigate the porphyrogenic potential of these selected
antihypertensive and analgesic drugs with CELCs: an inexpensive,
sensitive system that has been extensively used to study heme
metabolism and the porphyrogenic properties of many compounds (Granick,
1966; Tschudy and Bonkowsky, 1972; Schoenfeld et al., 1985; Marks et
al., 1986, 1987; Bonkovsky et al., 1992; Cable et al., 1994a; Hahn et
al., 1997). This system is analogous to the mammalian liver in vivo in
that it also maintains the inducibility and heme-dependent repression
of ALA synthase (Granick, 1966; Tschudy and Bonkowsky, 1972; Schoenfeld
et al., 1985; Bonkovsky et al., 1992; Cable et al., 1994a; Hahn et al.,
1997). The kinetics of heme synthesis in CELCs more closely resembles
that in human liver than in rodent models (Healey et al., 1981;
Bonkovsky et al., 1985). This system is also convenient for studying
test compounds over a wide range of doses, including low doses.
Antidepressant Agents.
Bupropion is an effective and usually
well tolerated antidepressant. An immediate release formulation has
been available in the United States since 1988, and a sustained-release
formulation was approved in 1996 (Settle, 1998). Because we were unable
to find any references to the porphyrogenic potential of bupropion, either in MEDLINE searches, literature references (Parikh and Moore,
1978; Eales, 1979; Moore, 1980; Moore and McColl, 1989; Dover et al.,
1994; Ashley, 1996; Moore and Hift, 1997; Gorchein, 1997; Kalman and
Bonkovsky, 1998), or at the University of Cape Town Web site
(www.uct.ac.za/depts/ liver/drugname.htm), we suggest that the data
presented in Fig. 1A are the first indication that bupropion can
increase porphyrin accumulation. We further suggest that treating acute
porphyric patients with bupropion be done with caution, and that other
antidepressants may pose a lower risk of exacerbating porphyric attacks.
Nefazodone was developed to treat major depression, and is generally
well tolerated (Dewan and Anand, 1999) even in high doses (Catalano et
al., 1999), although it has been reported to cause life-threatening
hepatocellular injury in an idiosyncratic manner (Aranda-Michel et al.,
1999). A MEDLINE search failed to produce any documents linking
nefazodone and porphyrin accumulation, but the University of Cape Town
database suggests that nefazodone should be used to treat acute
porphyric patients with extreme caution. Our results (Fig. 1B) indicate
that the combination of DES and the lowest concentration of nefazodone
tested (3.16 µM) significantly increases porphyrin accumulation
compared with DES alone. The combination of DES and 31.6 µM
nefazodone causes porphyrin accumulation in CELCs that is nearly as
great as the positive control (DES plus 2 mM PB). We propose that our
results, coupled with the recommendation from the University of Cape
Town, strongly suggest that the use of nefazodone to treat porphyric
patients may be more likely to exacerbate acute porphyric attacks than the use of some other antidepressants.
The results for fluoxetine presented in Fig. 2A, showing that combined
treatment of DES and fluoxetine does not result in increased porphyrin
accumulation in CELCs, are in general agreement with previous reports.
For example, Vaz and Salcedo (1991) reported the safe use of fluoxetine
in a 25-year-old woman with acute intermittent porphyria for treatment
of her depressive symptoms. Moore and Hift (1997) list fluoxetine as a
safe drug, and the database at the University of Cape Town recommends
that this drug be used with caution. Thus, it would appear that
fluoxetine is a reasonably safe drug for treating acute porphyric patients.
There is little information available concerning the porphyrogenic
potential of paroxetine, except in the database at the University of
Cape Town, which recommends that this drug be used with caution. Based
on the results presented in Fig. 2, neither paroxetine nor fluoxetine
increase porphyrin accumulation in CELCs, either with or without
concurrent treatment with DES. We suggest that these may be the
antidepressants of choice for use in acute porphyrias.
Benzodiazepine-Type Drugs.
In the present report, we evaluated
the porphyrogenic potential of five benzodiazepine-type drugs: oxazepam
(Fig. 3A), lorazepam (Fig. 3B), diazepam (Fig. 4A), triazolam (Fig.
4B), and midazolam (Fig. 5). When given in combination with DES, all
five of these drugs significantly increased porphyrin accumulation for
at least two of the concentrations tested. Diazepam, midazolam, and
triazolam produced significant increases even at the lowest
concentration tested (3.16 µM), whereas lorazepam and oxazepam
required higher concentrations (10 µM) to produce a significant
effect. Given these findings, lorazepam and oxazepam are probably
preferable to the others for the treatment of patients with acute porphyria.
Oxazepam has previously been reported to cause a small, but
statistically significant, increase in porphyrin accumulation in CELCs
at a concentration of 10 µg/ml, and in chick embryo liver in ovo (10 mg/egg) (Zimmer et al., 1980). It is variously listed as believed to be
safe (Kalman and Bonkovsky, 1998), unsafe (but with conflicting
results) (Moore, 1980; Moore and McColl, 1989; Moore and Hift, 1997),
or as "use with caution" (University of Cape Town database).
Despite the porphyrogenic result in CELCs (Fig. 3B), lorazepam is
reported to be safe for the treatment of patients with acute porphyria
(Moore and Hift, 1997; Moore and McColl, 1989; Dover et al., 1994) and
it is listed under the "Use" category in the University of Cape
Town database. Diazepam has been reported to increase porphyrin
accumulation in CELCs (Zimmer et al., 1980) and in chick embryo liver
in ovo (Zimmer et al., 1980; Deybach et al., 1987), but it does not
increase ALA synthase in rats (Parikh and Moore, 1978). Gorchein (1997) reports that diazepam has been used to treat porphyric patients, but it
is generally listed as being unsafe (Moore, 1980; Moore and McColl,
1989; Moore and Hift, 1997; Kalman and Bonkovsky, 1998) or contentious
(Ashley, 1996). Both triazolam and midazolam are generally listed as
safe (Moore and McColl, 1989; Dover et al., 1994; Ashley, 1996;
Gorchein, 1997; Moore and Hift, 1997).
In summary, although interspecies differences (particularly for drug
metabolism) and the difficulties of determining equivalent dosages
complicate the extrapolation of results from experimental models to
porphyric patients, the current studies suggest that patients with
acute porphyrias may be at greater risk for developing porphyric
attacks when treated with bupropion or nefazodone (compared with
fluoxetine or paroxetine), and that all of the benzodiazepine derivatives that were studied should be administered with caution to
such patients.
ALA, 5-aminolevulinate;
CELC, chick embryo
liver cell;
DES, desferrioxamine;
DMSO, dimethyl sulfoxide;
PB, phenobarbital.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
-aminolevulinate synthase from embryonic chick liver.
J Biol Chem
256:
9329-9333
Toxicogenetic diseases.
Cell Mol Biol
43:
89-94.
