Experimental Arthrogryposis Caused by Viral Myopathy Daniel B.
Drachman, MD; Leslie P. Weiner, MD; Donald
embryo has been postulated to cause the joint deformities in arthrogryposis multiplex congenita (AMC). Experimental damage to the motor neurons or pharmacologic \s=b\ Immobilization of the
blockade of neuromuscular transmission has previously resulted in typical joint changes of AMC. In the present investigation, we have studied the effects of paralysis produced by a viral myopathy on
joint development. Coxsackievirus A2 was injected intravenously into chick embryos on the seventh day of incubation. Within 48 hours, severe myositis and paralysis resulted. Electron microscopical and immunofluorescence techniques demonstrated virus in muscle cells. Within three to four days after infection, the muscle had virtually disappeared. Ankylosis of joints, corresponding to that seen in human AMC, occurred. This study shows that primary myopathy with paralysis can produce arthrogrypotic joint deformities. The possibility of a viral etiologic factor in some human cases of AMC should be considered. (Arch Neurol 33:362-367, 1976)
Arthrogryposis multiplex congenita (AMC) is a disorder character¬ ized by rigidity or "clubbing" of more than one joint, present at birth. It is not a single disease entity, but rather a syndrome of joint malformations, which usually occurs in a setting of neuromuscular disease. In hu¬ mans, arthrogryposis has been ob¬ served in association with lesions of the spinal cord1 and occasionally with congenital abnormalities of peripheral nerves2 or muscles.35
publication June 17, 1975. From the Department of Neurology, laboratories of neuromuscular diseases, neurovirology, and neuropathology, Johns Hopkins University
of the specific cause, affected infants have usually been partially paralyzed or immobilized before birth. The implication is that movement of the embrvo is essential for normal joint development.610 It has been postulated that the lack of embryonic movement per se may be the underlying factor responsible for the joint deformities.1115 To test this hypothesis experimentally, we have previously produced paralysis in chick embryos by removal of the spinal cord,9 or by the use of neuromuscular blocking agents.91015 In those experi¬ ments, interference with motor nerve function caused the paralysis that resulted in typical arthrogrypotic joint deformities. In the present study, we asked whether a primary disorder of muscle in the developing embryo might also lead to arthrogryposis. To resolve this question, we have utilized coxsackievirus A216 to produce infectious myositis with paralysis in chick embryos. This report describes the abnormali¬ ties of joint, muscle, and spinal cord that resulted, and their pathogenesis. Infection of the embryo with this common virus represents an instruc¬ tive model of congenital myopathy and arthrogryposis. MATERIALS AND METHODS Animals
Reprint requests to Department of Neurology, 317 Traylor Bldg, 1721 E Madison St, Baltimore, MD 21205
White leghorn chick embryos were main¬ tained at 37.7 C in a humidified, forceddraft incubator. There were 194 embryos used in these experiments. Animals used for virus preparation and assays were 1-day-old Swiss mice.
Virus
for
School of Medicine, Baltimore.
(Dr Drachman).
Price, MD; Janis Chase
Regardless
severe
Accepted
L.
In order to prepare a stock virus pool, coxsackievirus A2 (obtained from Amer¬ ican Type Tissue Culture Collection) was inoculated intramuscularly into newborn Swiss mice (0.025 ml). The mice were killed
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
48 hours
later,
at the onset of hind
leg paralysis. The skinned hind leg was homog¬ enized by Dulbecco phosphate buffered saline (10% weight by volume). The suspen¬ sion was centrifuged at 12,000 g for 15 minutes, and the supernatant was stored in ampules at -70 C. This stock virus pool contained IO8 5 newborn doses (LD50)/0.1 ml.
mouse
50% lethal
Inoculation Procedure Serial tenfold dilutions of the stock virus solution were made in chick Ringer solu¬ tion17 to final concentrations of 10" to 10" \ The solutions were injected directly into the chorioallantoic circulation of embryos by the following method, described pre¬ viously in detail1 s: a rectangular window was removed from the shell and shell membrane overlying the embryo to permit access for injection and observation. A specially designed polyethylene microcatheter was inserted within a chorioallantoic vein, under a binocular dissecting micro¬ scope. A micrometer-driven syringe at¬ tached to the catheter was used to make the injections. A wide range of embryonic ages and virus doses were tried; the most effective combination proved to be 10(i ·"* LD,0 of virus in 0.05-ml solution in sevenday chick embryos, and was therefore used in the majority of the experiments. After the injections, the openings in the shells were sealed with cellophane tape and the eggs were returned to the incubator. At one- to two-day intervals, the windows were opened and observations of em¬ bryonic movement were made under a -
dissecting microscope. Tissue
Preparation
After coxsackievirus injection (at 7 days' incubation), the embryos were killed at various times ranging from six hours to 13 days later. The embryonic tissues were prepared for the following: (1) light micros¬ copy, by fixation in isotonic 10% formolsaline; (2) electron microscopy, by immer¬ sion in a fixative composed of 3% glutaraldehyde, 1% paraformaldehyde, and 1% acrolein in 0.1 M cacodylate buffer (pH 7.4) with 3% sucrose; and (3) viral studies, by
freezing immediately on storing at —70 C until use.
solid C02 and
The formol-saline-fixed tissues were em¬ bedded in paraffin, sectioned semiserially at 6µ, and stained with hematoxylin-eosin. The tissues for electron microscopical studies were dehydrated in alcohols, em¬ bedded in Epon 812, and sectioned on an ultramicrotome. Thick sections were stained with toluidine blue; thin sections were stained with uranyl acetate and lead citrate and examined in an electron micro¬ scope.
Viral
Assay
Blood for virus assay was obtained by venipuncture of the chorioallantoic circula¬ tion, as described above. After the embryos were killed, the brain, spinal cord, heart, and skeletal muscle of the lower limbs were dissected free. To eliminate possible con¬ tamination of the exterior surfaces of the brain and spinal cord by virus-containing blood, they were washed thoroughly in cold Ringer solution, immersed in coxsackie A2 antiserum and then washed repeatedly in cold Ringer solution. The tissues were homogenized, and serial tenfold dilutions were made. Two litters of mice were inoc¬ ulated subcutaneously with 0.1 ml of each tissue homogenate dilution. The LD50 was calculated by the Reed-Muench method. 19
Fluorescent
Antibody
Antibody
Method
prepared by inoculating adult Swiss mice with 106 LD50 of coxsackievirus together with Freund complete adjuvant. Injections were repeated at tenday intervals. At the end of four weeks, the animals had accumulated ascitic fluid, which was removed with an 18-gauge needle. After processing according to the method of Larsen,20 the ascitic fluid was absorbed with acetone-prepared mouse muscle and bone powder to remove nonspe¬ cific antibodies. Chick embryo tissues were sectioned in a cryostat at 8µ and stained by the indirect method21 using anticoxsackie A2 antibody for 30 minutes, followed by goat antimouse IgG conjugated with fluorescein isothiocyanate. The stained tissues was
5
were
examined with
an
ultraviolet micro¬
scope.
RESULTS Within 24 to 48 hours after inocula¬ tion of IO6 5 LD50 of coxsackievirus into seven-day embryos, muscular move¬ ments ceased and remained absent for the rest of the incubation period. Embryos that received smaller doses, and those injected at earlier (day 3) or later (days 12 to 15) ages showed incomplete paresis. Normally, chick
embryos are quite active at this stage, making an average of 30 movements/ min at nine days.22 Gross
Pathologic Findings
The embryos showed striking musculoskeletal abnormalities, but other¬ wise continued to develop normally, apart from a moderate reduction in size (Fig 1, left). Limitation of joint movement was apparent within three to four days after inoculation, and firm joint fixation was present by 15 days' incubation age. The lower beak protruded beyond the upper in the coxsackie-treated embryos, an abnor¬ mality we previously noted in em¬ bryos paralyzed by neuromuscular blocking agents.10 The skeletal mus¬ cles appeared gelatinous in consis¬ tency within two to three days after virus injection, and subsequently di¬ minished strikingly in bulk. At later embryonic stages (17 days' incubation age and beyond), virtually no skeletal muscle remained. The spinal cords of the virus-treated chicks were smaller than normal. The embryos with in¬ complete paralysis due to modifica¬ tions of the schedules of viral inocula¬ tion (as described above) showed less severe muscle loss and joint restric¬ tion. ß
Histopathologic Findings Within one day after virus inocula¬ tion (eight days' incubation), the skel¬ etal muscle tissue showed evidence of active myopathy. Early changes af¬ fecting developing muscle included accumulations of dense amorphous material in the cytoplasm (Fig 2, left) and pyknosis of nuclei. Electron microscopy showed that the cytoplasm was filled with vesicles, granular and amorphous debris, and clusters of disoriented filaments (Fig 2, center). Later, the normal organelles could no longer be identified. The degenerative and inflammatory changes progressed rapidly; by three to five days after inoculation, the muscle was largely necrotic, and was undergoing phago¬ cytosis by inflammatory cells (Fig 2, right). At 13 to 15 days' incubation age, the muscle was almost entirely replaced by fat and fibrous connective tissue (Fig 1, right). Immunofluorescent staining
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
showed viral antigen in muscle 24 hours after viral inoculation (Fig 3, inset). Antigen could be detected by this method until muscle necrosis was far advanced. Electron microscopy at eight to ten days showed viral particles within degenerating muscle cells (Fig 3). Individual viral particles measured 18 to 20 nm with a centerto-center spacing of 23 to 25 nm. Some of the particles were within vesicles or adjacent to membranes; others were arranged in paracrystalline arrays (Fig 3, right). These viral particles are identical to those described by Harri¬ son et al.23 The knee and hock (ankle) joints were devoid of normal joint spaces. They were fused first by fibrous connective tissues, and subsequently
by cartilaginous bridges extending across the presumptive articular re¬ gions (Fig 1, center).
Abnormal motor neurons were first identified in the spinal cord early on the second day after inoculation. These degenerating cells were angu¬ lar in shape, with condensed nuclear chromatin and increased cytoplasmic density (Fig 4, upper left). Electron microscopy showed abundant free ribosomes in the cytoplasm, many of which were present in the form of ribosome crystals arranged in a tetrameric pattern (Fig 4, upper center, upper right). The tetramere were associated in planes or in three dimensional structures made up of stacked sheets. Identical ribosome crystals have been described24 in chick spinal neurons dying during normal development or after removal of the peripheral field of innervation. Viral particles could not be identified in the degenerating motor neurons, and immunofluorescent studies failed to dis¬ close viral antigen in the spinal cord. The degenerated motor neurons were quickly engulfed and removed by phagocytic cells. By 11 to 13 days, a reduction of the number of anterior horn cells was apparent, and by 17 to 20 days, there was notable depletion of motor neurons (Fig 4, lower left, lower right). Other parts of the spinal cord appeared relatively normal.
Virologie Findings In order to determine the
concen-
Fig 1.—Chick embryos injected with coxsackievirus A2 on day 7. Left, Fifteen-day embryo. Skin removed from leg, showing notable diminution of skeletal muscle; leg joints are rigid; upper beak is abnormally short. Center, Eighteen-day embryo. Knee joint, showing femur (F) and tibia (T) fused by cartilage bridging
presumptive joint region (hematoxylin-eosin, 20). Right, Eigh¬ teen-day embryo. Anterior thigh, showing that fat (asterisk) has replaced all muscle tissue; F indicates femur (hematoxylin-eosin, X50).
Fig 2.—Muscle of embryos infected with coxsackievirus A2 on day 7. Left, Eight-day embryo. Two myotubes (arrows) show early degenerative changes with accumulation of amorphous debris. Myotube (asterisk) and nerve (N) are normal (Epon 812embedded section stained with toluidine blue, 950). Center, Nine-day embryo. Electron micrograph shows muscle cell
containing necrotic debris (arrows) in form of vesicular, granular, and amorphous material. Adjacent myoblast (M) is normal; indicates nuclei (scale, 1µ; 6,300). Right, Fifteen-day embryo. Longitudinal section of muscle shows severe inflammatory myopathy with cellular infiltrate (hematoxylin-eosin, 225).
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
Fig
3.—Muscle of embryos infected on day 7 shows evidence of coxsackievirus. Left, Eight-day embryo. Electron micrograph of muscle (M) shows crystalline arrays of viral particles (arrows). Degenerative changes include presence of amorphous and gran¬ ular material, vesicles (V), lamellar bodies, and disoriented fila¬ ments (F); indicates nucleus (scale, 0.5µ; 32,000). Inset, Nineday embryo. Specific fluorescence (arrow) in skeletal muscle;
tration of coxsackievirus in several tissues of the chick embryos, infectivity studies were carried out by injections of tissue homogenates into newborn mice (Table). The highest concentration of virus was found in skeletal muscle within two days after
inoculation, reaching 7 LD50/0.1 gm. This finding correlated with both the strong fluorescent antibody stain¬ ing of viral antigen and the electron
observation of virions within the muscle cells. By seven days postinoculation, the muscle concentra¬ tion of virus had fallen to 2 2 LD50/ 0.1 gm, reflecting the notable loss of muscle tissue. The blood and heart contained lower concentrations of virus at all times, while the spinal cord levels were 105 6 LD50/0.1 gm. The data suggest that replication of virus occurred in the spinal cord, but neither fluorescent antibody staining nor electron microscopy demonstrated
microscopical
dark circles are muscle nuclei (x680). Right, Nine-day embryo. Electron micrograph shows coxsackievirus within muscle cyto¬ plasm. Viral particles are uniform in shape and size; they are round, measure 18 to 20 nm in diameter, have a center-to-center distance of 22 to 25 nm, and are organized in paracrystalline arrays (scale 0.1 µ; 104,000).
which cells within the spinal cord were infected. COMMENT The results of this study clearly show a consistent association between arthrogrypotic joint abnormalities and the destruction of muscle by coxsackievirus. The most profoundly paralyzed embryos showed severe joint abnormalities, while in embryos with incomplete paresis, lesser degree of joint restriction developed. The arthrogryposis was identical to that occurring in embryos paralyzed by neuromuscular blocking agents or by destruction of the spinal cord.9·1015 In the severe cases, joint cavities were absent, and fusion of the joints—first by fibrous connective tissue and later by cartilage-occurred. The abnormal¬ ly short upper beak was previously noted in chicks immobilized with botulinum toxin,10 and it is a consequence of malposition and impaired growth
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
of the skeletal elements secondary to the absence of muscular contraction. Skeletal muscle clearly bore the brunt of the coxsackievirus infection. In the severely involved embryos, muscle necrosis occurred within one to two days after inoculation, with virtually complete destruction of skel¬ etal muscle within four to six days. Our findings in skeletal muscle con¬ firm and extend those of Peers et al.25 The presence of virus within muscle was shown by electron microscopy, immunofluorescence, and viral infectivity studies. The degree of muscle involvement and subsequent joint deformity de¬ pended on the timing and dose of virus inoculation. The maximal effect was achieved by injection of a rela¬ tively large dose of coxsackievirus at seven days' incubation. Why the muscle was more susceptible to coxsackieinfection at days 7 than at
earlier
later
developmental ages is not yet clear; perhaps the difference is or
due to the stage of maturation of skeletal muscle, or to other factors that might be important in host resistance to viral infections. Less complete paresis occurred in embryos
injected at relatively early (day 3) or late (days 12 to 15) stages, and those given smaller virus inocula. In these embryos, fewer joints were involved, the joint restriction was less pro¬ nounced, and there was no evidence of joint fusion or shortening of the upper beak.
Although coxsackie
ent in the brains and
virus
was
pres¬
spinal cords
of infected embryos, the muscle disease cannot be attributed to primary neu¬ ronal involvement. First, the coxsackie-treated embryos showed notable muscle disease at an earlier stage than occurs as a result of denervation.6 Second, the degree of muscle destruc¬ tion and inflammation was far more severe than is ever seen with denerva¬ tion.26·27
Whether the loss of anterior J^orn cells in spinal cords of experimental embryos was due to coxsackievirus infection per se or was a secondary effect resulting from the muscle damage with resultant loss of the "pe¬ ripheral field of innervation"28 cannot be definitely distinguished on the basis of the present material. At no time was there a cellular response in the spinal cord, although the ability of coxsackievirus to provoke an inflam¬ matory response in the chick embryo was proven by the striking histopathologic changes in skeletal muscle. Electron microscopy and immunofluorescence studies did not disclose viral particles or antigens in the motor neurons or other spinal cord cells. However, infectivity studies demon¬ strated that the concentration of virus in the spinal cord and brain was 50 times greater than that in the blood, suggesting that viral replication oc¬ curred within the central nervous system. The appearance of ribosome crystals within dying neurons unfor-
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
Fig 4.—Spinal cords of chick embryos, showing ventral horns. Top left, Nine-day embryo. Necrotic debris is all that remains of two neurons (center and lower left). Dark neuron (arrow) contains abundance of ribosomes in cytoplasm. Several nor¬ mal-appearing motor neurons are present at left of micrograph. Note lack of inflam¬ mation (Epon 812-embedded section stained with toluidine blue; 700). Top center, top right, Nine-day embryo. Elec¬ tron micrographs of dark neuron similar to that shown at arrow in 4, upper left. Cyto¬ plasm is filled with free ribosomes and ribosomal crystals. Top center, Crystals are composed of four ribosomes in square tetramers. Upper right, Stacks of tetramers seen on edge (scale 0.1 µ, 50,000). Bottom left, Control 15-day embryo. Ven¬ tral horn of lumbar spinal cord is normal. Note abundance of motor neurons (hema¬ toxylin-eosin, x225). Bottom right, Fif¬ teen-day embryo. Infected with coxsackie¬ virus on day 7. Ventral horn of lumbar spinal cord. Note paucity of motor neurons and lack of inflammation (hematoxylineosin, x225).
tunately does not clarify this problem
further because an identical pattern of crystals has been found under a variety of conditions including hypo¬ thermia29 and death of supernumerary cells in normal development.24 Since ribosome crystals form in neurons degenerating after loss of their pe¬ ripheral field of innervation,24 it is tempting to speculate that the motor neuron changes in the present experi¬ ment represent secondary effects fol¬ lowing the destruction of muscle by the coxsackie infection. Alternatively, the crystal formation might be caused by direct infection of neurons by Coxsackievirus A2, due to a different mechanism. It is well known that one
Coxsackie Virus Concentrations* in Chick Embryo Tissues Incubation Age When
Killed, Days
Tissuef Blood Heart Brain
Spinal
13 13 13 13 20
20 20 20
cord Muscle Blood Heart Brain Muscle Blood Brain Spinal cord
Muscle
0.1 gm 104.5 104.1 105.9 105.6 107.1 1023 103.3 104.3 106.3 102.3 103.6 102.9 1022
*
Virus concentrations determined by method of Reed and Muensch.i9 t Virus injections on day 7 in all embryos. $ Indicates 50% lethal doses.
of the earliest manifestations of picornavirus infections is an inhibition of host cell RNA and protein synthesis.30 This biochemical change could lead to an abundance of free ribosomes, which might then aggregate into tetramers.
The joints did not show evidence of viral infection or inflammation as judged by immunofluorescence and histologie criteria. Thus, it is most likely that the joint ankylosis resulted from the immobilization of the em¬ bryo, produced in this case by viral myositis. The observation that a my¬
opathy
can
produce typical joint
deformities adds further support to the concept that paralysis and immo¬ bilization are important factors in the pathogenesis of AMC. How closely does the coxsackie model correspond to naturally occur¬ ring AMC in humans? In terms of articular pathologic findings, the coxsackie-infected embryos showed fixa¬ tion of joints in postures dictated by the position of the embryo within the egg, without loss of any of the bony elements; this clearly fits the defini¬ tion of AMC. The joint fixation in the severely infected chicks was more marked than that usually seen in human cases of AMC, because the degree of neuromuscular involvement was more profound than is present in most human cases. Indeed, such severe destruction of muscle is not
compatible
with survival to term in humans or in the chick. However, in the less severely affected chick em¬ bryos, the soft-tissue ankylosis of scattered joints, particularly the hock and foot joints, corresponds closely to that seen in the human disease. Most human cases of AMC occur sporadically, and in the large major¬ ity, the cause remains unknown.1 The possibility that a prenatal viral infec¬ tion of the nervous system or muscles may exist in some cases has been raised in the past on the basis of suggestive, but not definitive evi¬ dence.14·31·32 The present work pro¬ vides an experimental model of AMC due to infection with a virus that commonly produces a mild illness in humans. It is well established that infection with some viruses may produce far more severe changes in embryonic than in adult animals,33 and a coxsackievirus A9 has recently been isolated from muscle of a child with a congenital myopathy.34 One implication of our study is that viruses may possibly be of etiologic impor¬ tance in human cases of AMC. This investigation was supported by grants 5R01 HD04817-05, 5-POl-NS 10920-02, and NS 10580-01A1. Dr Weiner is a recipient of research career development grant NS 50274 from the Public Health Service.
References 1. Drachman DB:
Congenital
deformities produced by neuromuscular disorders of the developing embryo, in Norris FH, Kurland LT (eds): Motor Neuron Diseases. New York, Grune & Stratton Inc, 1969. 2. Bargeton E, Nezelef C, Guran P, et al: \l=E'\tude anatomique d'un cas d'arthrogrypose multiplex congenitale et familiale. Rev Neurol 104:479-489, 1961. 3. Banker BQ, Victor M, Adams RD: Arthrogryposis multiplex due to congenital muscular dystrophy. Brain 80:319-334, 1957. 4. Pearson CM, Fowler WG: Hereditary nonprogressive muscular dystrophy inducing arthrogryposis syndrome. Brain 86:75-88, 1963. 5. Gubbay SS, Walton JN, Pearce GW: Clinical and pathological study of a case of congenital muscular dystrophy. J Neurol Neurosurg Psychiatry 29:500-520, 1966. 6. Murray PDF, Selby D: Intrinsic and extrinsic factors in the primary development of the skeleton. Wilhelm Roux Arch Entwicklungsmechanik Organ 122:627-662, 1930. 7. Hamburger V, Waugh M: The primary development of the skeleton in nerveless and poorly innervated limb transplants of chick embryos. Physiol Zool 13:367-380, 1940. 8. Badgley CE: Correlation of clinical and anatomical facts leading to a conception of the etiology of congenital hip dysplasias. J Bone
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
Joint Surg 25:503-523, 1943. 9. Drachman DB, Sokoloff L: The role of movement of embryonic joint development. Dev Biol 14:401-420, 1966. 10. Murray PDF, Drachman DB: The role of movement in the development of joints and related structures: The head and neck in the chick embryo. J Embryol Exp Morphol 22:349-371, 1969. 11. Gilmour JR: Amyoplasia congenita. J Pathol Bacteriol 58:675-685, 1946. 12. Brandt S: Arthrogryposis multiplex congenita. Acta Paediatr 34:365-381, 1947. 13. Whittem JH: Congenital abnormalities in calves: Arthrogryposis and hydranencephaly. J Pathol Bact 73:375-387, 1957. 14. Drachman DB, Banker BQ: Arthrogryposis multiplex congenita. Arch Neurol 5:77-93, 1961. 15. Drachman DB, Coulombre AJ: Experimental clubfoot and arthrogryposis multiplex congenita. Lancet 2:523-526, 1962. 16. Huebner RJ, Ranson SE, Beeman EA: Studies of coxsackie virus. Public Health Rep 65:803-806, 1950. 17. Rugh R: Laboratory Manual of Vertebrate Embryology, ed 5. Minneapolis, Burgess, 1961. 18. Drachman DB, Coulombre AJ: Method for continous infusion of fluids into the chorioallantoic circulation of the chick embryo. Science 138:144-145, 1962. 19. Reed LJ, Muensch HA: A simple method of estimating 50% endpoints. Am J Hyg 27:493-497, 1938. 20. Larson VM: Preparation of fluorescein labeled coxsackie A14 antibody from immune mouse ascitic fluid. Virology 18:497-500, 1962. 21. Goldman M: Fluorescent Antibody Methods. New York, Academic Press, 1968, pp 154\x=req-\ 157. 22. Drachman DB: The developing motor endplate: Pharmacological studies in the chick embryo. J Physiol 169:707-712, 1963. 23. Harrison AK, Murphy FA, Gary GW Jr: Ultrastructural pathology of coxsackie A4 virus infection of mouse striated muscle. Exp Mol Pathol 14:30-42, 1971. 24. O'Connor TM, Wyttenbach CR: Cell death in the embryonic spinal cord. J Cell Biol 60:448\x=req-\ 459, 1974. 25. Peers JH, Ransom SE, Huebner RJ: The pathological changes produced in chick embryos by yolk sac inoculation of group A coxsackie virus. J Exp Med 96:17-26, 1952. 26. Eastlick HL: Studies on transplanted embryonic limbs of the chick. J Exp Zool 93:27-49, 1943. 27. Drachman DB: Is acetylcholine the trophic neuromuscular transmitter? Arch Neurol 17:206\x=req-\ 218, 1967. 28. Price DL: The influence of the periphery on spinal motor neurons. Ann NY Acad Sci 228:355\x=req-\ 363, 1976. 29. Byers B: Structure and formation of ribosome crystals in hypothermic chick embryo cells. J Mol Biol 26:155-167, 1964. 30. Holland JJ: Enterovirus entrance into specific host cells and subsequent alterations of cell protein and nucleic acid synthesis. Bacteriol Rev 28:3-13, 1964. 31. Gregg NM: The occurrence of congenital defects in children following maternal rubella. Med J Aust 2:122-126, 1945. 32. Rudolph AJ, Yow MD, Phillips CA, et al: Transplacental rubella infection in newly born infants. JAMA 191:843-845, 1965. 33. Johnson RT: Effects of viral infection on the developing nervous system. N Engl J Med 287:599-604, 1972. 34. Tang TT, Sedmak GV, Siegesmund KA, et al: Chronic myopathy associated with coxsackie type A9. N Engl J Med 292:608-611, 1975.
Drachman, MD; Leslie P. Weiner, MD; Donald
embryo has been postulated to cause the joint deformities in arthrogryposis multiplex congenita (AMC). Experimental damage to the motor neurons or pharmacologic \s=b\ Immobilization of the
blockade of neuromuscular transmission has previously resulted in typical joint changes of AMC. In the present investigation, we have studied the effects of paralysis produced by a viral myopathy on
joint development. Coxsackievirus A2 was injected intravenously into chick embryos on the seventh day of incubation. Within 48 hours, severe myositis and paralysis resulted. Electron microscopical and immunofluorescence techniques demonstrated virus in muscle cells. Within three to four days after infection, the muscle had virtually disappeared. Ankylosis of joints, corresponding to that seen in human AMC, occurred. This study shows that primary myopathy with paralysis can produce arthrogrypotic joint deformities. The possibility of a viral etiologic factor in some human cases of AMC should be considered. (Arch Neurol 33:362-367, 1976)
Arthrogryposis multiplex congenita (AMC) is a disorder character¬ ized by rigidity or "clubbing" of more than one joint, present at birth. It is not a single disease entity, but rather a syndrome of joint malformations, which usually occurs in a setting of neuromuscular disease. In hu¬ mans, arthrogryposis has been ob¬ served in association with lesions of the spinal cord1 and occasionally with congenital abnormalities of peripheral nerves2 or muscles.35
publication June 17, 1975. From the Department of Neurology, laboratories of neuromuscular diseases, neurovirology, and neuropathology, Johns Hopkins University
of the specific cause, affected infants have usually been partially paralyzed or immobilized before birth. The implication is that movement of the embrvo is essential for normal joint development.610 It has been postulated that the lack of embryonic movement per se may be the underlying factor responsible for the joint deformities.1115 To test this hypothesis experimentally, we have previously produced paralysis in chick embryos by removal of the spinal cord,9 or by the use of neuromuscular blocking agents.91015 In those experi¬ ments, interference with motor nerve function caused the paralysis that resulted in typical arthrogrypotic joint deformities. In the present study, we asked whether a primary disorder of muscle in the developing embryo might also lead to arthrogryposis. To resolve this question, we have utilized coxsackievirus A216 to produce infectious myositis with paralysis in chick embryos. This report describes the abnormali¬ ties of joint, muscle, and spinal cord that resulted, and their pathogenesis. Infection of the embryo with this common virus represents an instruc¬ tive model of congenital myopathy and arthrogryposis. MATERIALS AND METHODS Animals
Reprint requests to Department of Neurology, 317 Traylor Bldg, 1721 E Madison St, Baltimore, MD 21205
White leghorn chick embryos were main¬ tained at 37.7 C in a humidified, forceddraft incubator. There were 194 embryos used in these experiments. Animals used for virus preparation and assays were 1-day-old Swiss mice.
Virus
for
School of Medicine, Baltimore.
(Dr Drachman).
Price, MD; Janis Chase
Regardless
severe
Accepted
L.
In order to prepare a stock virus pool, coxsackievirus A2 (obtained from Amer¬ ican Type Tissue Culture Collection) was inoculated intramuscularly into newborn Swiss mice (0.025 ml). The mice were killed
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
48 hours
later,
at the onset of hind
leg paralysis. The skinned hind leg was homog¬ enized by Dulbecco phosphate buffered saline (10% weight by volume). The suspen¬ sion was centrifuged at 12,000 g for 15 minutes, and the supernatant was stored in ampules at -70 C. This stock virus pool contained IO8 5 newborn doses (LD50)/0.1 ml.
mouse
50% lethal
Inoculation Procedure Serial tenfold dilutions of the stock virus solution were made in chick Ringer solu¬ tion17 to final concentrations of 10" to 10" \ The solutions were injected directly into the chorioallantoic circulation of embryos by the following method, described pre¬ viously in detail1 s: a rectangular window was removed from the shell and shell membrane overlying the embryo to permit access for injection and observation. A specially designed polyethylene microcatheter was inserted within a chorioallantoic vein, under a binocular dissecting micro¬ scope. A micrometer-driven syringe at¬ tached to the catheter was used to make the injections. A wide range of embryonic ages and virus doses were tried; the most effective combination proved to be 10(i ·"* LD,0 of virus in 0.05-ml solution in sevenday chick embryos, and was therefore used in the majority of the experiments. After the injections, the openings in the shells were sealed with cellophane tape and the eggs were returned to the incubator. At one- to two-day intervals, the windows were opened and observations of em¬ bryonic movement were made under a -
dissecting microscope. Tissue
Preparation
After coxsackievirus injection (at 7 days' incubation), the embryos were killed at various times ranging from six hours to 13 days later. The embryonic tissues were prepared for the following: (1) light micros¬ copy, by fixation in isotonic 10% formolsaline; (2) electron microscopy, by immer¬ sion in a fixative composed of 3% glutaraldehyde, 1% paraformaldehyde, and 1% acrolein in 0.1 M cacodylate buffer (pH 7.4) with 3% sucrose; and (3) viral studies, by
freezing immediately on storing at —70 C until use.
solid C02 and
The formol-saline-fixed tissues were em¬ bedded in paraffin, sectioned semiserially at 6µ, and stained with hematoxylin-eosin. The tissues for electron microscopical studies were dehydrated in alcohols, em¬ bedded in Epon 812, and sectioned on an ultramicrotome. Thick sections were stained with toluidine blue; thin sections were stained with uranyl acetate and lead citrate and examined in an electron micro¬ scope.
Viral
Assay
Blood for virus assay was obtained by venipuncture of the chorioallantoic circula¬ tion, as described above. After the embryos were killed, the brain, spinal cord, heart, and skeletal muscle of the lower limbs were dissected free. To eliminate possible con¬ tamination of the exterior surfaces of the brain and spinal cord by virus-containing blood, they were washed thoroughly in cold Ringer solution, immersed in coxsackie A2 antiserum and then washed repeatedly in cold Ringer solution. The tissues were homogenized, and serial tenfold dilutions were made. Two litters of mice were inoc¬ ulated subcutaneously with 0.1 ml of each tissue homogenate dilution. The LD50 was calculated by the Reed-Muench method. 19
Fluorescent
Antibody
Antibody
Method
prepared by inoculating adult Swiss mice with 106 LD50 of coxsackievirus together with Freund complete adjuvant. Injections were repeated at tenday intervals. At the end of four weeks, the animals had accumulated ascitic fluid, which was removed with an 18-gauge needle. After processing according to the method of Larsen,20 the ascitic fluid was absorbed with acetone-prepared mouse muscle and bone powder to remove nonspe¬ cific antibodies. Chick embryo tissues were sectioned in a cryostat at 8µ and stained by the indirect method21 using anticoxsackie A2 antibody for 30 minutes, followed by goat antimouse IgG conjugated with fluorescein isothiocyanate. The stained tissues was
5
were
examined with
an
ultraviolet micro¬
scope.
RESULTS Within 24 to 48 hours after inocula¬ tion of IO6 5 LD50 of coxsackievirus into seven-day embryos, muscular move¬ ments ceased and remained absent for the rest of the incubation period. Embryos that received smaller doses, and those injected at earlier (day 3) or later (days 12 to 15) ages showed incomplete paresis. Normally, chick
embryos are quite active at this stage, making an average of 30 movements/ min at nine days.22 Gross
Pathologic Findings
The embryos showed striking musculoskeletal abnormalities, but other¬ wise continued to develop normally, apart from a moderate reduction in size (Fig 1, left). Limitation of joint movement was apparent within three to four days after inoculation, and firm joint fixation was present by 15 days' incubation age. The lower beak protruded beyond the upper in the coxsackie-treated embryos, an abnor¬ mality we previously noted in em¬ bryos paralyzed by neuromuscular blocking agents.10 The skeletal mus¬ cles appeared gelatinous in consis¬ tency within two to three days after virus injection, and subsequently di¬ minished strikingly in bulk. At later embryonic stages (17 days' incubation age and beyond), virtually no skeletal muscle remained. The spinal cords of the virus-treated chicks were smaller than normal. The embryos with in¬ complete paralysis due to modifica¬ tions of the schedules of viral inocula¬ tion (as described above) showed less severe muscle loss and joint restric¬ tion. ß
Histopathologic Findings Within one day after virus inocula¬ tion (eight days' incubation), the skel¬ etal muscle tissue showed evidence of active myopathy. Early changes af¬ fecting developing muscle included accumulations of dense amorphous material in the cytoplasm (Fig 2, left) and pyknosis of nuclei. Electron microscopy showed that the cytoplasm was filled with vesicles, granular and amorphous debris, and clusters of disoriented filaments (Fig 2, center). Later, the normal organelles could no longer be identified. The degenerative and inflammatory changes progressed rapidly; by three to five days after inoculation, the muscle was largely necrotic, and was undergoing phago¬ cytosis by inflammatory cells (Fig 2, right). At 13 to 15 days' incubation age, the muscle was almost entirely replaced by fat and fibrous connective tissue (Fig 1, right). Immunofluorescent staining
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
showed viral antigen in muscle 24 hours after viral inoculation (Fig 3, inset). Antigen could be detected by this method until muscle necrosis was far advanced. Electron microscopy at eight to ten days showed viral particles within degenerating muscle cells (Fig 3). Individual viral particles measured 18 to 20 nm with a centerto-center spacing of 23 to 25 nm. Some of the particles were within vesicles or adjacent to membranes; others were arranged in paracrystalline arrays (Fig 3, right). These viral particles are identical to those described by Harri¬ son et al.23 The knee and hock (ankle) joints were devoid of normal joint spaces. They were fused first by fibrous connective tissues, and subsequently
by cartilaginous bridges extending across the presumptive articular re¬ gions (Fig 1, center).
Abnormal motor neurons were first identified in the spinal cord early on the second day after inoculation. These degenerating cells were angu¬ lar in shape, with condensed nuclear chromatin and increased cytoplasmic density (Fig 4, upper left). Electron microscopy showed abundant free ribosomes in the cytoplasm, many of which were present in the form of ribosome crystals arranged in a tetrameric pattern (Fig 4, upper center, upper right). The tetramere were associated in planes or in three dimensional structures made up of stacked sheets. Identical ribosome crystals have been described24 in chick spinal neurons dying during normal development or after removal of the peripheral field of innervation. Viral particles could not be identified in the degenerating motor neurons, and immunofluorescent studies failed to dis¬ close viral antigen in the spinal cord. The degenerated motor neurons were quickly engulfed and removed by phagocytic cells. By 11 to 13 days, a reduction of the number of anterior horn cells was apparent, and by 17 to 20 days, there was notable depletion of motor neurons (Fig 4, lower left, lower right). Other parts of the spinal cord appeared relatively normal.
Virologie Findings In order to determine the
concen-
Fig 1.—Chick embryos injected with coxsackievirus A2 on day 7. Left, Fifteen-day embryo. Skin removed from leg, showing notable diminution of skeletal muscle; leg joints are rigid; upper beak is abnormally short. Center, Eighteen-day embryo. Knee joint, showing femur (F) and tibia (T) fused by cartilage bridging
presumptive joint region (hematoxylin-eosin, 20). Right, Eigh¬ teen-day embryo. Anterior thigh, showing that fat (asterisk) has replaced all muscle tissue; F indicates femur (hematoxylin-eosin, X50).
Fig 2.—Muscle of embryos infected with coxsackievirus A2 on day 7. Left, Eight-day embryo. Two myotubes (arrows) show early degenerative changes with accumulation of amorphous debris. Myotube (asterisk) and nerve (N) are normal (Epon 812embedded section stained with toluidine blue, 950). Center, Nine-day embryo. Electron micrograph shows muscle cell
containing necrotic debris (arrows) in form of vesicular, granular, and amorphous material. Adjacent myoblast (M) is normal; indicates nuclei (scale, 1µ; 6,300). Right, Fifteen-day embryo. Longitudinal section of muscle shows severe inflammatory myopathy with cellular infiltrate (hematoxylin-eosin, 225).
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
Fig
3.—Muscle of embryos infected on day 7 shows evidence of coxsackievirus. Left, Eight-day embryo. Electron micrograph of muscle (M) shows crystalline arrays of viral particles (arrows). Degenerative changes include presence of amorphous and gran¬ ular material, vesicles (V), lamellar bodies, and disoriented fila¬ ments (F); indicates nucleus (scale, 0.5µ; 32,000). Inset, Nineday embryo. Specific fluorescence (arrow) in skeletal muscle;
tration of coxsackievirus in several tissues of the chick embryos, infectivity studies were carried out by injections of tissue homogenates into newborn mice (Table). The highest concentration of virus was found in skeletal muscle within two days after
inoculation, reaching 7 LD50/0.1 gm. This finding correlated with both the strong fluorescent antibody stain¬ ing of viral antigen and the electron
observation of virions within the muscle cells. By seven days postinoculation, the muscle concentra¬ tion of virus had fallen to 2 2 LD50/ 0.1 gm, reflecting the notable loss of muscle tissue. The blood and heart contained lower concentrations of virus at all times, while the spinal cord levels were 105 6 LD50/0.1 gm. The data suggest that replication of virus occurred in the spinal cord, but neither fluorescent antibody staining nor electron microscopy demonstrated
microscopical
dark circles are muscle nuclei (x680). Right, Nine-day embryo. Electron micrograph shows coxsackievirus within muscle cyto¬ plasm. Viral particles are uniform in shape and size; they are round, measure 18 to 20 nm in diameter, have a center-to-center distance of 22 to 25 nm, and are organized in paracrystalline arrays (scale 0.1 µ; 104,000).
which cells within the spinal cord were infected. COMMENT The results of this study clearly show a consistent association between arthrogrypotic joint abnormalities and the destruction of muscle by coxsackievirus. The most profoundly paralyzed embryos showed severe joint abnormalities, while in embryos with incomplete paresis, lesser degree of joint restriction developed. The arthrogryposis was identical to that occurring in embryos paralyzed by neuromuscular blocking agents or by destruction of the spinal cord.9·1015 In the severe cases, joint cavities were absent, and fusion of the joints—first by fibrous connective tissue and later by cartilage-occurred. The abnormal¬ ly short upper beak was previously noted in chicks immobilized with botulinum toxin,10 and it is a consequence of malposition and impaired growth
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
of the skeletal elements secondary to the absence of muscular contraction. Skeletal muscle clearly bore the brunt of the coxsackievirus infection. In the severely involved embryos, muscle necrosis occurred within one to two days after inoculation, with virtually complete destruction of skel¬ etal muscle within four to six days. Our findings in skeletal muscle con¬ firm and extend those of Peers et al.25 The presence of virus within muscle was shown by electron microscopy, immunofluorescence, and viral infectivity studies. The degree of muscle involvement and subsequent joint deformity de¬ pended on the timing and dose of virus inoculation. The maximal effect was achieved by injection of a rela¬ tively large dose of coxsackievirus at seven days' incubation. Why the muscle was more susceptible to coxsackieinfection at days 7 than at
earlier
later
developmental ages is not yet clear; perhaps the difference is or
due to the stage of maturation of skeletal muscle, or to other factors that might be important in host resistance to viral infections. Less complete paresis occurred in embryos
injected at relatively early (day 3) or late (days 12 to 15) stages, and those given smaller virus inocula. In these embryos, fewer joints were involved, the joint restriction was less pro¬ nounced, and there was no evidence of joint fusion or shortening of the upper beak.
Although coxsackie
ent in the brains and
virus
was
pres¬
spinal cords
of infected embryos, the muscle disease cannot be attributed to primary neu¬ ronal involvement. First, the coxsackie-treated embryos showed notable muscle disease at an earlier stage than occurs as a result of denervation.6 Second, the degree of muscle destruc¬ tion and inflammation was far more severe than is ever seen with denerva¬ tion.26·27
Whether the loss of anterior J^orn cells in spinal cords of experimental embryos was due to coxsackievirus infection per se or was a secondary effect resulting from the muscle damage with resultant loss of the "pe¬ ripheral field of innervation"28 cannot be definitely distinguished on the basis of the present material. At no time was there a cellular response in the spinal cord, although the ability of coxsackievirus to provoke an inflam¬ matory response in the chick embryo was proven by the striking histopathologic changes in skeletal muscle. Electron microscopy and immunofluorescence studies did not disclose viral particles or antigens in the motor neurons or other spinal cord cells. However, infectivity studies demon¬ strated that the concentration of virus in the spinal cord and brain was 50 times greater than that in the blood, suggesting that viral replication oc¬ curred within the central nervous system. The appearance of ribosome crystals within dying neurons unfor-
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
Fig 4.—Spinal cords of chick embryos, showing ventral horns. Top left, Nine-day embryo. Necrotic debris is all that remains of two neurons (center and lower left). Dark neuron (arrow) contains abundance of ribosomes in cytoplasm. Several nor¬ mal-appearing motor neurons are present at left of micrograph. Note lack of inflam¬ mation (Epon 812-embedded section stained with toluidine blue; 700). Top center, top right, Nine-day embryo. Elec¬ tron micrographs of dark neuron similar to that shown at arrow in 4, upper left. Cyto¬ plasm is filled with free ribosomes and ribosomal crystals. Top center, Crystals are composed of four ribosomes in square tetramers. Upper right, Stacks of tetramers seen on edge (scale 0.1 µ, 50,000). Bottom left, Control 15-day embryo. Ven¬ tral horn of lumbar spinal cord is normal. Note abundance of motor neurons (hema¬ toxylin-eosin, x225). Bottom right, Fif¬ teen-day embryo. Infected with coxsackie¬ virus on day 7. Ventral horn of lumbar spinal cord. Note paucity of motor neurons and lack of inflammation (hematoxylineosin, x225).
tunately does not clarify this problem
further because an identical pattern of crystals has been found under a variety of conditions including hypo¬ thermia29 and death of supernumerary cells in normal development.24 Since ribosome crystals form in neurons degenerating after loss of their pe¬ ripheral field of innervation,24 it is tempting to speculate that the motor neuron changes in the present experi¬ ment represent secondary effects fol¬ lowing the destruction of muscle by the coxsackie infection. Alternatively, the crystal formation might be caused by direct infection of neurons by Coxsackievirus A2, due to a different mechanism. It is well known that one
Coxsackie Virus Concentrations* in Chick Embryo Tissues Incubation Age When
Killed, Days
Tissuef Blood Heart Brain
Spinal
13 13 13 13 20
20 20 20
cord Muscle Blood Heart Brain Muscle Blood Brain Spinal cord
Muscle
0.1 gm 104.5 104.1 105.9 105.6 107.1 1023 103.3 104.3 106.3 102.3 103.6 102.9 1022
*
Virus concentrations determined by method of Reed and Muensch.i9 t Virus injections on day 7 in all embryos. $ Indicates 50% lethal doses.
of the earliest manifestations of picornavirus infections is an inhibition of host cell RNA and protein synthesis.30 This biochemical change could lead to an abundance of free ribosomes, which might then aggregate into tetramers.
The joints did not show evidence of viral infection or inflammation as judged by immunofluorescence and histologie criteria. Thus, it is most likely that the joint ankylosis resulted from the immobilization of the em¬ bryo, produced in this case by viral myositis. The observation that a my¬
opathy
can
produce typical joint
deformities adds further support to the concept that paralysis and immo¬ bilization are important factors in the pathogenesis of AMC. How closely does the coxsackie model correspond to naturally occur¬ ring AMC in humans? In terms of articular pathologic findings, the coxsackie-infected embryos showed fixa¬ tion of joints in postures dictated by the position of the embryo within the egg, without loss of any of the bony elements; this clearly fits the defini¬ tion of AMC. The joint fixation in the severely infected chicks was more marked than that usually seen in human cases of AMC, because the degree of neuromuscular involvement was more profound than is present in most human cases. Indeed, such severe destruction of muscle is not
compatible
with survival to term in humans or in the chick. However, in the less severely affected chick em¬ bryos, the soft-tissue ankylosis of scattered joints, particularly the hock and foot joints, corresponds closely to that seen in the human disease. Most human cases of AMC occur sporadically, and in the large major¬ ity, the cause remains unknown.1 The possibility that a prenatal viral infec¬ tion of the nervous system or muscles may exist in some cases has been raised in the past on the basis of suggestive, but not definitive evi¬ dence.14·31·32 The present work pro¬ vides an experimental model of AMC due to infection with a virus that commonly produces a mild illness in humans. It is well established that infection with some viruses may produce far more severe changes in embryonic than in adult animals,33 and a coxsackievirus A9 has recently been isolated from muscle of a child with a congenital myopathy.34 One implication of our study is that viruses may possibly be of etiologic impor¬ tance in human cases of AMC. This investigation was supported by grants 5R01 HD04817-05, 5-POl-NS 10920-02, and NS 10580-01A1. Dr Weiner is a recipient of research career development grant NS 50274 from the Public Health Service.
References 1. Drachman DB:
Congenital
deformities produced by neuromuscular disorders of the developing embryo, in Norris FH, Kurland LT (eds): Motor Neuron Diseases. New York, Grune & Stratton Inc, 1969. 2. Bargeton E, Nezelef C, Guran P, et al: \l=E'\tude anatomique d'un cas d'arthrogrypose multiplex congenitale et familiale. Rev Neurol 104:479-489, 1961. 3. Banker BQ, Victor M, Adams RD: Arthrogryposis multiplex due to congenital muscular dystrophy. Brain 80:319-334, 1957. 4. Pearson CM, Fowler WG: Hereditary nonprogressive muscular dystrophy inducing arthrogryposis syndrome. Brain 86:75-88, 1963. 5. Gubbay SS, Walton JN, Pearce GW: Clinical and pathological study of a case of congenital muscular dystrophy. J Neurol Neurosurg Psychiatry 29:500-520, 1966. 6. Murray PDF, Selby D: Intrinsic and extrinsic factors in the primary development of the skeleton. Wilhelm Roux Arch Entwicklungsmechanik Organ 122:627-662, 1930. 7. Hamburger V, Waugh M: The primary development of the skeleton in nerveless and poorly innervated limb transplants of chick embryos. Physiol Zool 13:367-380, 1940. 8. Badgley CE: Correlation of clinical and anatomical facts leading to a conception of the etiology of congenital hip dysplasias. J Bone
Downloaded From: http://archneur.jamanetwork.com/ by a Penn State Milton S Hershey Med Ctr User on 05/23/2015
Joint Surg 25:503-523, 1943. 9. Drachman DB, Sokoloff L: The role of movement of embryonic joint development. Dev Biol 14:401-420, 1966. 10. Murray PDF, Drachman DB: The role of movement in the development of joints and related structures: The head and neck in the chick embryo. J Embryol Exp Morphol 22:349-371, 1969. 11. Gilmour JR: Amyoplasia congenita. J Pathol Bacteriol 58:675-685, 1946. 12. Brandt S: Arthrogryposis multiplex congenita. Acta Paediatr 34:365-381, 1947. 13. Whittem JH: Congenital abnormalities in calves: Arthrogryposis and hydranencephaly. J Pathol Bact 73:375-387, 1957. 14. Drachman DB, Banker BQ: Arthrogryposis multiplex congenita. Arch Neurol 5:77-93, 1961. 15. Drachman DB, Coulombre AJ: Experimental clubfoot and arthrogryposis multiplex congenita. Lancet 2:523-526, 1962. 16. Huebner RJ, Ranson SE, Beeman EA: Studies of coxsackie virus. Public Health Rep 65:803-806, 1950. 17. Rugh R: Laboratory Manual of Vertebrate Embryology, ed 5. Minneapolis, Burgess, 1961. 18. Drachman DB, Coulombre AJ: Method for continous infusion of fluids into the chorioallantoic circulation of the chick embryo. Science 138:144-145, 1962. 19. Reed LJ, Muensch HA: A simple method of estimating 50% endpoints. Am J Hyg 27:493-497, 1938. 20. Larson VM: Preparation of fluorescein labeled coxsackie A14 antibody from immune mouse ascitic fluid. Virology 18:497-500, 1962. 21. Goldman M: Fluorescent Antibody Methods. New York, Academic Press, 1968, pp 154\x=req-\ 157. 22. Drachman DB: The developing motor endplate: Pharmacological studies in the chick embryo. J Physiol 169:707-712, 1963. 23. Harrison AK, Murphy FA, Gary GW Jr: Ultrastructural pathology of coxsackie A4 virus infection of mouse striated muscle. Exp Mol Pathol 14:30-42, 1971. 24. O'Connor TM, Wyttenbach CR: Cell death in the embryonic spinal cord. J Cell Biol 60:448\x=req-\ 459, 1974. 25. Peers JH, Ransom SE, Huebner RJ: The pathological changes produced in chick embryos by yolk sac inoculation of group A coxsackie virus. J Exp Med 96:17-26, 1952. 26. Eastlick HL: Studies on transplanted embryonic limbs of the chick. J Exp Zool 93:27-49, 1943. 27. Drachman DB: Is acetylcholine the trophic neuromuscular transmitter? Arch Neurol 17:206\x=req-\ 218, 1967. 28. Price DL: The influence of the periphery on spinal motor neurons. Ann NY Acad Sci 228:355\x=req-\ 363, 1976. 29. Byers B: Structure and formation of ribosome crystals in hypothermic chick embryo cells. J Mol Biol 26:155-167, 1964. 30. Holland JJ: Enterovirus entrance into specific host cells and subsequent alterations of cell protein and nucleic acid synthesis. Bacteriol Rev 28:3-13, 1964. 31. Gregg NM: The occurrence of congenital defects in children following maternal rubella. Med J Aust 2:122-126, 1945. 32. Rudolph AJ, Yow MD, Phillips CA, et al: Transplacental rubella infection in newly born infants. JAMA 191:843-845, 1965. 33. Johnson RT: Effects of viral infection on the developing nervous system. N Engl J Med 287:599-604, 1972. 34. Tang TT, Sedmak GV, Siegesmund KA, et al: Chronic myopathy associated with coxsackie type A9. N Engl J Med 292:608-611, 1975.
