694 SHIGELLA 



toxic substances were precipitated by alcohol and ether, and largely destroyed by heat 

 at 75° C. Todd (1904) in this country was able to show that 4 to 6 weeks' cultures of 

 Shiga's bacillus contained a soluble toxin, which was highly active on rabbits and horses, 

 but to a much less extent on guinea-pigs, rats, and mice. Flexner's bacillus proved incapable 

 of giving rise to a soluble toxin. Dopter (1905), studying the histological appearances of 

 rabbits dying of paralysis subsequent to injection with 24-hours' broth cultures of Sh. 

 shigce, observed definite lesions in the spinal cord, consisting chiefly of chromatolysis of 

 the anterior horn cells, sometimes with small interstitial haemorrhages and focal necroses 

 of the grey matter. The lesions occurred as frequently after the injection of toxin as of 

 bacilli. Further work by Kraus and Dorr ( 1905) led to the conclusion that Shiga's bacillus 

 gave rise to two toxins : (1) a soluble toxin, present in 8-10-days' broth cultures and in 

 filtered saline suspensions of 24-hours' agar cultures ; this was fatal to rabbits but not to 

 guinea-pigs, and gave rise to the production of a specific neutrahzing antitoxin ; (2) an 

 insoluble toxin present in the bacterial bodies, which was fatal both to rabbits and to guinea- 

 pigs. No soluble toxins were found in cultures of Flexner's bacillus. 



Flexner and Sweet (1906), using a modification of Conradi's method, obtained a toxin 

 from 24-hours' agar cultures of Shiga's bacillus, which, injected intravenously into rabbits, 

 gave rise to diarrhoea, paralysis, convulsions, and death. The paralysis began in the upper 

 limbs and extended at times to the lower limbs. Sometimes the animals survived for 10 

 d^s after the extremities were paralysed ; they lay on one side in a position of opisthotonos. 

 Post mortem, small haemorrhages were seen in the brain, and softening of the grey matter 

 in the spinal cord. In the intestine there was congestion of the serosa ; the walls of the 

 gut, especially of the caecum and appendix, were thickened and oedematous, the mucosa 

 was yellowish-white and thrown into deep folds, which were sometimes covered by a 

 pseudo-membrane or stippled with haemorrhages. The mesenteric glands were swoUen, 

 oedematous, and congested. Heat at 81° C. for 1 hour destroyed the toxin. These observa- 

 tions were largely confirmed by Bessau (1911), who concluded that there were two dififerent 

 toxins— one a paretic or neuro-toxin causing paralysis of the muscles, the other a marasmic 

 or intestinal toxin causing a fall in temperature, diarrhoea, and chronic marasmus. The 

 paretic toxin was neutralized by antitoxin, the marasmic toxin was not. Further, rabbits 

 were affected by both toxins, whereas in guinea-pigs the paretic toxin was without 

 effect. 



Bessau's conclusions were supported by the observations of Ohtsky and Kligler (1920), 

 who concluded that Sh. shigce formed two toxins : one a neurotoxin acting on the central 

 nervous system of the rabbit and identified by Ohtsky and KHgler as an exotoxin ; the 

 other having a specific affinity for the intestine and regarded by Ohtsky and Khgler as 

 an endotoxin. 



This conception was challenged by Okell and Blake (1930), who found that toxin 

 was not Hberated from the cell bodies in the absence of autolysis, and therefore concluded 

 that there was only one toxin produced, namely an endotoxin. The more recent chemical 

 work, however, of Boivin and Mesrobeanu {19Bla-h, 1938) leaves little doubt that the 

 dual conception is correct. By means of the trichloracetic acid method these workers 

 were able to demonstrate the presence in cultures of Shiga's baciUus of (1) a thermostable 

 glycolipoid somatic antigen, which they regarded as an endotoxin having an enterotoxic 

 effect, and (2) a thermolabUe protein substance, which they regarded as an exotoxin 

 having a neurotropic effect. The exotoxin is specific to Shiga's bacillus, and may be 

 formed by either antigenically smooth or rough strains of this organism (Haas 1937, 

 Istrati 1938, Steabben 1943, Olitzki et al. 1943) ; the endotoxin is similar to the smooth 

 somatic antigens found in other members of the dysentery group (Haas 1938a, h, c) and 

 in the Salmonella group (see p. 724), and is absent from antigenically rough strains. 



It is perhaps unfortunate that the terms endotoxin and exotoxin have been used to 

 refer to the two types of toxic substances produced by Shiga's bacfilus. The so-called 

 exotoxin is closely bound up with the cell bodies and, as the observations of Okell and 

 Blake (1930) showed, (see also OHtzki, Bendersky and Koch 1943), is not excreted 



