close

Вход

Забыли?

вход по аккаунту

?

Cytochrome c oxidase reaction improves histopathological assessment of zidovudine myopathy.

код для вставкиСкачать
Cytochrome c Oxldase Reaction
Improves Histopathological Assessment
of Zidovudine Myopathy
Patrick Chariot, MD,+t Isabelle Monnet, MD,$ and Romain Gherardi, MDX
Zidovudine can induce a mitochondrial myopathy with ragged-red fibers and partial cytochrome c oxidase deficiency.
In an attempt to improve histological assessment of zidovudine myopathy, we evaluated cytochrome c oxidase histochemical reaction in the muscle of 10 patients with biopsy-proven zidovudine myopathy (Group l), 10 myopathic
zidovudine receivers without typical histopathological features of zidovudine myopathy (Group 2), and 10 human
immunodeficicncy virus (HIVI-infected patients not treated by zidovudine who had an immunohistological profile of
HIV-associated myopathy or other neuromuscular disorders (Group 3). Among zidovudine receivers, cytochrome c
oxidase deficiency was found in 10 of 10 patients from Group 1 and 7 of 10 from Group 2. No cytochrome c oxidase
deficiency was observed in patients not treated by zidovudine. When present, cytochrome c oxidase-negative fibers
accounted for 2 to 28% of fibers, and there was no difference for the number of cytochrome c oxidase-negative fibers
between Group 1 and Group 2. Most patients with cytochrome c oxidase deficiency that could be evaluated clinically
after muscle biopsy improved after withdrawal of zidovudine (5 of 7 in Group 1 , 5 of 5 in Group 2). Patients who did
not improve had an HIV-associated myopathy concurrently with zidovudine myopathy. We conclude that cytochrome
c oxidase reaction may be used as a reliable marker of zidovudine mitochondrial toxicity in HIV-infected patients
with muscular symptoms.
Chariot P, Monnet I, Gherardi R. Cytochrome c oxidase reaction improves histopathological
assessment of zidovudine myopathy. Ann Neurol 1993;34:561-565
Zidovudine is the first-line antiretroviral therapy in
patients with acquired immunodeficiency syndrome
(AIDS) [ l ] .It can induce a mitochondrial myopathy
with ragged-red fibers [2-41. However, some pathologists are reluctant to accept the diagnosis of zidovudine
myopathy [5}. Ragged-red fibers may be rare and usually combine mitochondrial accumulation with myofibritlar loss, which may hinder their recognition [3]. In
a number of patients, mitochondrial changes are mixed
with inflammatory changes 131 and, on some occasions,
the polymyositic process of human immunodeficiency
virus (H1V)-associated myopathy {b}overwhelms zidovudine myopathy. Moreover, muscle biopsy of patients
with muscular symptoms receiving low-dose zidovudine may be unremarkable, lacking typical features of
zidovudine myopathy [?I. This suggests that histological expression of mitochondrial damage could be delayed, as characteristic lesions occur in patients with
heavy cumulated doses of zidovudine [3, S}.
Mitochondrial dysfunction with impaired cytochrome c oxidase (CCO) activity [ 3 , 4 ] and histochemical evidence of a partial CCO deficiency in skeletal
muscle [4, S} have been demonstrated in patients with
From the Departments of *Pathology (Neuropathology) and Woxicology, HBpital Henri Mondor, and $Department of Pneumology,
Centre Hospitalier Intercommunal de Cr6tei1, Crete& France.
full-blown zidovudine myopathy, suggesting that the
CCO reaction could be appropriate for the detection
of muscular toxicity of zidovudine. Partial CCO deficiency has been commonly described in genetically induced mitochondrial myopathies 193 and has been occasionally detected in inflammatory myopathies { 101.
The present study evaluated the CCO reaction for
the detection of zidovudine mitochondrial muscular
toxicity and was performed to answer the following
questions: (1) Is CCO deficiency demonstrable in
zidovudine recipients with muscular symptoms in the
absence of histological evidence of zidovudine m y o p
athy? (2) May CCO deficiency be found in HIVinfected patients not treated by zidovudine? And (3),
is there any clinical evidence that myopathy with partial
CCO deficiency in HIV-infected patients is related to
zidovudine ?
Patients
We studied three h-oups of H1V-infected patients, as follows; Group 1 was 10 consecutive patients with typical zidovudine myopathy; Group 2 was 10 consecutive myopathic
patients treated with zidovudine but without histopathologi-
Received Feb 19, 1793, and in revised form Apr 21. Accepted for
publication May 12, 1993.
Address correspondence to Dr Chariot, Dkparternent de Pathologie
(Neuropathologie),HBpital Henri Mondor, 94 000 Crbteil, France.
Copyright 0 1993 by the American Neurological Association
561
cal features of zidovudine myopathy; and Group 3 was 10
consecutive HIV-infected patients with neuromuscular
symptoms who did not receive zidovudine. None of these
30 patients were included in previous morphological studies
on zidovudine myopathy by our group [ 3 , 81.
and M H C class I1 (HLA-DR: Becton Dickinson, Mountain
View, CA) antigens.
In each case, CCO activity was evaluated in at least 200
fibers and classified as present or absent. In fibers with only
segmental enzyme deficiency [ll], we considered that CCO
activity was present. Endomysial inflammation was assessed
semiquantitatively and quoted as absent, mild, moderate, or
marked. The HIV-associated myopathy was recognized when
the immunohistological profile of polymyositis was present
{b) and was defined by the combination of endomysial inliltrates of CD8-positive cells expressing M H C class I1 antigens
with sarcolemmal expression of MHC class I antigens in a
majority of muscle fibers. Zidovudine myopathy was assessed
by the association of ragged-red fibers on trichrome and myofilamentous abnormalities including cytoplasmic body formation. The clinical courses after biopsy of patients were reviewed with particular attention paid to follow-up after
withdrawal of zidovudine.
Student's t test and the determination of the correlation
coefficient r were used for statistical analysis.
Methods
Muscle specimens were frozen in liquid nitrogen-cooled isopentane ( - 160°C) and stored at - 80°C. Serial 8-hm-thick
cross sections were stained with conventional dyes (hematoxylin and eosin, modified Gomori trichrome, periodic acid
Schiff, and oil red 0)and processed for the following histochemical reactions: ATPase (at p H 9.4, 4.6, and 4.31,
NADH-tetrazolium reductase, succinate dehydrogenase, and
CCO. Immunocytochemistry was performed on frozen material using an immunoalkaline phosphatase to identify T8 lymphocytes (CD8: Leu-2a), T 4 lymphocytes and macrophages
(CD4:Leu-3), macrophages (CD14: Leu-M3) (Becton Dickinson, Mountain View, CA), major histocompatibility locus
(MHC) class I (HLA-ABC: Dakopatts, Glostrup, Denmark),
Clinical Data and Pathological Findings
cco
Inflammation
Patient Age (yr)!Sex
RRF Fibers Withdrawal
Endomysial Septal (9%) (9%) of Zidovudine
Total Zidovudine
Intake (gm)
Histological Diagnosis
Clinical Symptoms
Patients with zidovudine myopathy
1 34/M
Proximal weakness
2 30/M
Myalgias
3 33/M
Myalgias
4 481M
Myalgias
5 28/M
Myalgias
6 33/M
Muscle weakness
7 641M
Myalgias, proximal weakness
8 28/M
Muscle weakness
9 32/F
Myalgias, proximal weakness
10 68lM
Proximal weakness
470
400
360
230
230
590
650
500
1,500
200
AZT-My
AZT-My
AZT-My
AZT-My
AZT-My
AZT-My
AZT-My
An-My
AZTMy, HlV-My
AZT-My, HlV-My
0
+
++
+
+
+
t
+
I+
+++
Zidovudine receivers without histological evidence of zidovudine myopathy
11 35!M
Mvalgias. myopathic EMG
300
12 311M
Myalgias
300
13 3U/M
300
Myakm
14 50/M
Myalgias, proximal weakness
120
15 36!M
Proximal weakness
500
16 35lF
880
Myalg~as,proximal weakness
17 32!M
Assymetric motor deficit
90
18 44!M
450
Myalgias
19 25lF
Myalgias, slim disease
320
20 26/F
Myalgias, myopathic EMG
60
Minimal changes
0
Minimal changes
0
Minimal changes
0
+
Minimal changes
+
HIV-My
Minimal changes
0
Minimal changes
0
Toxoplasma-associated myoritis +
Diffuse atrophy
0
I t
Vasculitis
HIV-infected patients
21
32!M
22 47!M
23 46!M
24 24!M
25 26lM
26 78lF
27
30lM
28 55lM
29 29lM
30 74!M
HIV-My
HIV-My
HIV-My
HIV-My
Minimal changes
Minimal changes
Minimal changes
Minimal changes
Minimal changes
Vascditis
not treated hy zidovudine
Myalgias, atrophy
Myalgias
Predominantly neuropathic features
Predominantly neuropathic features
Myalgias
Predominantly neumpathic features
Predominantly neuropathic feacures
Predominantly neuropathic features
Myalgias
Predominantly neuropathic features
0
0
0
0
0
0
0
0
0
0
0
+
++ t
++
+
+
+
+
+
++
+
t
0
+
12
28
6
17
8
2
3
1
1
12
9
11
5
12
26
Improvement
Improvement
Improvement
Improvement
Improvement
N o t evaluable'
N o t evaluableh
N o t evaluableb
Worsening
Worsening
17
s
3
14
4
2
6
0
0
0
Improvement
Improvement
Improvement
Improvement
Improvement
N o t done
Not evaluablrb
N o change
N o change
Improvement'
0
0
0
0
++
+
0
0
+
++
0
0
0
0
0
+
+
+
+++
+
+
+
+
0
+
+
0
0
0
0
+
+++
+
++
+
0
I
+
+
+
0
0
0
0
0
0
9
7
2
6
17
0
0
0
...
...
...
0
0
0
...
0
...
...
...
0
0
...
...
o . . .
Zidovudine myopathy (AZT-My) was defined as the association of mitochondrial changes with ragged-red fibers (RRF), myohlamentous abnormalities, and cytoplasmic bodies on muscle biopsy. T h e human immunodeficiency virus (HIV)-associated myopathy (HIV-My) was recognized when the immunohisrological profile
of polymyositis was present and was defined by the combination of endomysial infiltrates of CD8-positive cells expressing major histocompatibility locus (MHC)
class 11 antigens with sarcolemmal expression of M H C class I a n t i e n s in a majority of muscle fibers. Inflammation was assessed semiquantitatively and quoted as
absent (O), mild ( + 1, moderate ( + +), or marked ( + +).
+
"HIV-wasting syndrome developed subsequently.
bl>eath occurred shortly after zidovudine wlthdrawal.
'Onset of myalgias occurred before zidowdine therapy was starred. Parrial improvement followed zidovudine withdrawal.
CCO fibers
=
muscle fibers negativeiy stained by cytochrome F oxidase histochemical reaction; EMG
562 Annals of Neurology Vol 34 No 4 October 1993
=
electromyogram
Fig 1. Zidovudine myopathy. (A) Ragged-red fibws with
muvked myofkzmentous changes typical of zidwudine myopathy.
(Patient 2, cryostat section, trachrome, X 450 before 8% reduction.) (B) Focal cytochrome c oxidase drficiency in the same patient. (Patient 2, cvyostat section, cytochrome c oxidase, x 450
before 8% reduction.)
Results
Main results are presented in the Table. In patients
from Group 1, ragged-red fibers accounted for 1 to
1296 of fibers (mean, 6%). In this group, myofilamentous abnormalities were observed in all cases and
included reduction in the myofibrillar green on trichrome, fibers with punch-out myofibrillar loss, and
constant cytoplasmic body formation (Fig 1A). The
polyrnyositic process of HIV-associated myopathy was
Fig 2. Human immunode$ciency virus-associated myopathy.
{A) Endomysial infiltrate composed of CD8-positive cells. (Patient 24, cryostat section, immunoalkuline phosphatase reaction,
x 280 bdore 8% reduction.) (B) Nomull cytochrome c oxidase
activity in the same patient. Note that type 1 jbers have a
higher cytocbmme c oxidase activity than type 2 fibers. {Patient
24, cryostat section, cytochrome c oxidase, x 280 before 8% reduction.)
recognized in 7 patients (2 patients from Group 1, 1
patient from Group 2, and 4 patients from Group 3 )
(Fig 2A). Endomysial inflammatory infiltrates without
sarcolemmal expression of MHC class I antigens in a
majority of muscle fibers were found in 15 patients (7
patients from Group 1, 3 patients from Group 2, and
5 patients from Group 3). Septal inflammation was observed in 26 patients, including 2 with vasculitis.
A partial CCO deficiency was found in all patients
Chariot et al: CCO Deficiency and Zidovudine
563
Fig 3. Focal cytochrome c oxidase deficiency in a zidouudine receiver witb unremarkable muscle histology. (Patient 13, cryostat
section, cytochrome c ariduse, x 280 before 38% reduction.)
with the typical histological pattern of zidovudine myopathy (Group 1) (Fig 1B). In Group 2, 7 patients of
10 had a partial CCO deficiency (Figs 3, 4). In Group
3, no CCO-negative fibers were found (Fig 2B). The
CCO-negative fibers were of both types 1 and 2 and
accounted for 2 to 28% of fibers (mean, 11%). A majority of CCO-negative fibers appeared normal on trichrome. All ragged-red fibers and all fibers with cytoplasmic bodies were CCO negative.
In Group 1, 5 of 7 patients improved after withdrawal of zidovudine. The 2 remaining patients had an
HIV-associated myopathy and their condition worsened. In Group 2, clinical improvement after withdrawal of zidovudine was observed in 6 of 8 patients
who could be evaluated. The 2 remaining patients, who
did not improve after withdrawal of zidovudine, had
no CCO-negative fibers. N o correlation was found between the percentage of CCO-negative fibers and the
cumulative dose of zidovudine. No significant differences were found between Groups 1 and 2 for the
percentage of CCO-negative fibers and for the cumulative dose of zidovudine.
Discussion
In the present study, CCO deficiency was found in
patients with full-blown zidovudine myopathy as well
as in myopathic zidovudine recipients without histological changes characteristic of zidovudine myopathy. No
CCO deficiency was observed in patients not treated
by zidovudine. Most patients with CCO deficiency improved after withdrawal of zidovudine. Patients who
did not improve had a concurrent muscle disorder
related to AIDS, such as polymyositis, i.e., HIVassociated myopathy [GI, or cachexia consistent with
the HIV-wasting syndrome [12).
Cytochrome c oxidase, the complex IV of mitochondrial respiratory chain, is the terminal enzyme of the
electron transport chain and consists of 13 subunits,
3 of which are encoded by mitochondrial DNA
564 Annals o f Neurology Vol 34 N o 4 October 1993
Fig 4. Zidwudine receiver with unremarkable muscle histology.
(A, B) Serial sections showing that a cytochrome c oxidase negative fiber has no abnomlity in succinate dehydmgenaJe reaction. (Patient 15, cryastat sections, (A) cytochrome c oxidase,
(B) succinate dehydmgenase, x 450 before 8% reduction.)
(mtDNA) and 10 by nuclear DNA [13]. Zidovudine
is a thymidine analogue that cannot form 3’phosphodiester linkages and thus terminates DNA
elongation when incorporated into a growing DNA
strand [14]. In patients with zidovudine myopathy, a
selective fall in the activity of enzymes partially encoded by mtDNA 131 and rntDNA depletion f l 5 ]
have been found, and mtDNA inhibition was proposed
as the mechanism of zidovudine toxicity. The present
study confirmed our previous finding of a focal CCO
deficiency in zidovudine myopathy [S], a rather puz-
zling finding if one considers the easy penetration of
zidovudine in all muscle fibers by passive, nonfacilitated diffusion across cell membranes {16}. The distribution and rates of CCO-negative fibers were similar
to what may be observed in genetically determined
mitochondrial myopathies related to mtDNA abnormalities { 111. In these conditions, the mosaic of cells
with different mitochondrial enzymatic capacities is
thought to reflect random cytoplasmic segregation of
abnormal mtDNA during cell divisions at developmental stages [17}. In normal individuals, no fibers
with CCO-negative phenotype can be detected despite
the presence of a variable number of natural CCOnegative mitochondria, mixed with CCO-positive mitochondria and distributed in a mosaic pattern [77. The
number of CCO-deficient mitochondria must reach a
threshold level in a cell to transform the cell phenotype from CCO positive to CCO negative {9}. The
focal pattern of zidovudine-associated CCO deficiency
might be related to such a threshold effect. The number of mitochondria with zidovudine-induced CCO
deficiency might increase in all muscle fibers, and
cells containing a high proporuon of natural CCOdeficient mitochondria might first switch to the CCOnegative phenotype, giving an apparently focal pattern
of CCO deficiency. Patient 20 improved partially after
zidovudine withdrawal but had no CCO deficiency at
muscle biopsy. This might be due to an early mmifestation of mitochondrial toxicity, undetectable with the
CCO reaction because the proportion of affected mitochondria may have been too small to transform muscle
cell phenotype from CCO positive to CCO negative.
The absence of correlation between the percentage of
CCO-negative fibers and the cumulative dose of zidovudine is puzzling but might reflect interindividual
variations in the baseline mitochondrial function, either
genetically induced or acquired, such as variations relative to tissular hypoxia [ls}. Only 1 patient in Group
1 and 2 patients in Group 2 had no inflammatory infiltrates in their muscle. Studies on zidovudine muscular
toxicity in animals and in cultured human muscle cells
have yielded mixed results [17-22), and the ability of
zidovudine to induce a myopathy in the absence of
HIV infection is still uncertain {22).
We conclude that CCO deficiency is found in patients with muscular manifestations related to zidovudine, either with or without typical histopathological
findings at muscle biopsy, and is not detected in patients not treated by zidovudine, including those with
HIV-associated myopathy. Histochemical reaction for
CCO may therefore be a reliable marker of zidovudine
mitochondrial toxicity in HIV-infected patients.
This work was supported by grants to Dr Chariot from the Association Fragaise contre les Myopathies and to Drs Chariot and Gherardi from the Agence Nationale de Recherches sur le SlDA and
by a European Concerted Action on “Neuropathology of AIDS.”
References
1. Johnson VA. New developments in antiretroviral drug therapy
for HIV infection. In: Volberding P, Jacobson MA, eds. AIDS
clinical review. New York: Marcel Dekker, 1992:69-104
2. Dalakas MC, Illa I, Pezeshkpour G H , et al. Mitochondrial myopathy caused by long-term zidovudine therapy. N Engl J Med
1990;322:1098-1105
3. Mhiri C,Baudrimorit M, Bonne G, et al. Zidovudine myopathy:
a distinctive disorder associated with mitochondrial dysfunction.
Ann Neurol 1991;29:606-614
4. Tomelleri G, Tonin P, Spadaro M, et al. AZT-induced mitochondrial myopathy. Ital J Neurol Sci 1992;13:723-728
5. Simpson DM, Wolfe DE. Neuromuscular complications of HIV
infection and its treatment. AIDS 1991;5:917-926
6. Illa I, Nath A, Dalakas M. Immunocytochemical and virological
characteristics of HIV-associated inflammatory myopathies: similarities with seronegative polymyositis. Ann Neurol 1991;29:
474-481
7. Jay C, Ropka M, Hench K, et al. Prospective study of myopathy
during prolonged low-dose AZT: Clinical correlates of AZT
mitochondrial myopathy and HIV-associated inflammatory myopathy. Neurology 1992;42(suppl 3):145 (Abstract)
8. Chariot P,Gherardi R. Partial cytochrome c oxidase deficiency
and cytoplasmic bodies in patients with zidovudine myopathy.
Neuromusc Disord 1991;1:357-363
9. Nonaka I, Koga Y ,l k u c h i A, Goto Y . Mitochondtial encephalomyopdthies and cytochrome c oxidase deficiency: muscle culture study. Acm Neuropathol 1991;82:286-294
10. Carpenter S, Karpati G, Johnston W, et al. Coexistence of polymyositis with mitochondrial myopathy. Neurology 1992;42
(suppl 1):188(Abstract)
11. Johnson MA, Bindoff LA, Turnbull DM. Cytochrome c oxidase
activiry in single muscle fibers: assay techniques and diagnostic
applications. Ann Neurol 1993;33:28-35
12. Belec L, Mhiri C , Di Costanzo B, Gherardi R. The HIV wasting
syndrome. muscle Nerve 1992;15:856-857
13. Cooper CE, Nicholls P, Freedman JA. Cytochrorne oxidase:
structure, function, and membrane topology of the polypeptde
subunits. Biochem Cell Biol 1991;69:586-607
14. Yarchoan R, Mitsuya H, Myers CE, Broder S. Clinical pharma(zidovudine) and recology of 3‘-azido-2’,3‘-dideoxythymidine
lated dideoxynucleosides. N Engl J Med 1989;321:726-738
15. Arnaudo E, Dalakas iM,Shanske S, et al. Depletion of muscle
mirochondrial DNA in AIDS patients with zidovudine-induced
myopathy. Lancet 1991;337:508-5 10
16. Zimmermann TP, Mahony WB, Prus KL. 3’-Azido-3’-deoxythymidine: an unusual nucleoside analogue that permeates the
membrane of human erythrocytes and lymphocytes by nonfacilitated diffusion. J Biol Chem 1987;262:5748-5754
17. Linnane AW, Ozawa T, Marzuki S, Tanaka M. Mitochondrial
DNA mutations as an important contributor to ageing and degenerative diseases. Lancet 1989;1:642-64 5
18. Corral-Debrinski M, Stepien G, Shoffner JM, et al. Hypoxemia
is associated with mitochondrial damage and gene induction.
JAMA 1991;266:1812-1816
19. Lamperth L, Dalakas MC, Dagani F, et al. Abnorma! skeletal
and cardiac muscle mitochondria induced by zidovudine (AZT)
in human muscle in vitro and in an animal model. Lab Invest
1991;65:742-751
20. Lewis W, Gonzalez B, Chomyn A, Papoian T. Zidovudine induces molecular, biochemical, and ultrastructural changes in rat
skeletal muscle mitochondria. J Clin Invest 1992;89:1354-1360
21. Herzherg N H , %om I , Zwart R, et al. hlajor growth reduction
and minor decrease in mitochondria! enzyme activity in cultured
human muscle cells after exposure to zidovudine. Muscle Nerve
192;15:706-7 10
22. Reyes MG, Casanova J, Varricchio F, et al. Zidovudine rnyopathy. Neurology 1992:42: 1252
Chariot e t al: CCO Deficiency and Zidovudine
565
Документ
Категория
Без категории
Просмотров
1
Размер файла
645 Кб
Теги
myopathy, cytochrome, zidovudine, reaction, histopathological, assessment, improve, oxidase
1/--страниц
Пожаловаться на содержимое документа