Search

Vesikari T., Ruuska T., Delem A., Andre F. E., Beards G. M. and Flewett T. H. (Department of Clinical Sciences, University of Tampere, Tampere, Finland, and the Biological Division, Smith Kline RIT Rixensart, Belgium and WHO Collaborating Centre for Reference and Research on Human Rotaviruses, East Birmingham Hospital, Birmingham, U.K.). Efficacy of two doses of RIT 4237 bovine rotavirus vaccine for prevention of rotavirus diarrhoea. Acta Paediatr Scand 80: 173, 1991. Candidate oral bovine rotavirus vaccine RIT 4237 or placebo was given to 252 Finnish infants at birth and at 7 months of age. No vaccine-associated reactions were observed. Primary rotavirus ELISA IgM responses were detected in 36% of the infants after the first vaccination; after the second dose 68% of the vaccinees were seropositive for rotavirus ELISA IgG antibody. The infants remained in clinical follow-up over two rotavirus epidemic seasons (total 28 months). Counted from child years in follow-up the overall vaccine protection rate was 43%. The clinical severity of rotavirus episodes was assessed using numerical score 0–20. Vaccine protection rate for cases with score ≥7 was 57% and for cases with score ≥11 it was 89%. It is concluded that vaccination with bovine rotavirus vaccine at birth and at 7 months of age, with the second dose given shortly before rotavirus epidemic season, protects infants against moderately severe and severe rotavirus diarrhoea in the first 2 years of life.

Rotaviruses (duoviruses) can be found in more than half the cases of acute diarrhoea in children up to the age of six or seven. About that age almost everyone has antibodies to them. Second infections occur and may not be as rare as laboratory findings so far suggest. Very young infants sometimes get subclinical disease-the effect of maternal antibody transmitted across the placenta? Very similar viruses, all possessing a common antigen detectable by immunofluorescence, are known to infect and/or cause diarrhoea in children, calves, piglets, mice, foals and monkeys. The calf virus and the human virus both infect piglets; piglet virus infects calves; we don't know whether any of these can infect children. Other mammals probably have similar diarrhoea viruses. An antigen common to all these viruses is probably in the inner capsid layer, and "species-specific" antigens are probably in the outer capsid layer. A precise test for comparing different strains is bably needed. Adenoviruses possibly cause a smaller proportion of cases of diarrhoea. Coronaviruses, well-known as enteric pathogens of pigs and calves, appear also to infect adults and children. 27 nm particles and 22-23 nm particles of density 1.4 (and other particles) can be found in faeces of children with natural diarrhoea and adults with experimental diarrhoea, sometimes in enormous numbers. It is not yet established whether they cause disease. Rotaviruses, animal coronaviruses and "Norwalk" virus attack the disaccharidase-producing epithelium of the small bowel; adenovirus pathology is unknown.A safe attenuated live vaccine strain of the human rotavirus urgently needs to be developed.

It is well established that rotaviruses of group A are the most important cause of severe diarrhoea in children. The causes of most cases of infectious diarrhoea still remain unidentified, however, and there must be other viruses to be found. 'Novel' rotaviruses have recently been discovered, mainly in animals (serogroups B and C in pigs and humans, D in birds, and one or more further groups in sheep and other mammals). Except for the group B virus which has caused widespread outbreaks of quite severe diarrhoea in adults in China (still not reported from outside China) these novel rotaviruses are rarities in the human and probably represent uncommon zoonotic infections. We speculate that the Chinese virus might have arisen by reassortment of genetic segments of animal group B viruses or perhaps by mutation, and so became infectious for man. The problem of identifying and determining the importance of small round viruses is reviewed. It seems likely that the group of small, round structured viruses, including Norwalk and the viruses of plainly calicivirus morphology, are all representatives of a whole group of enteric caliciviruses. Until they can be more easily cultivated it will be difficult to make diagnostic reagents available to all. Astroviruses and genuine parvoviruses have been found by many people in many countries. Astroviruses are probably more important as pathogens in lambs than in children or calves; parvoviruses can only be established as significant in epidemics. The coronavirus-like particles, first found in Vellore and Bristol, are still enigmatic and their role in diarrhoea is uncertain. Toroviridae, recently discovered as causes of diarrhoea in ungulates, do not seem to be at all important as causes of diarrhoea in humans. Possibly fruitful approaches to future searches are outlined: firstly to make more extensive use of immuno-electron microscopy; and secondly to try to improve existing tissue culture systems to make them more sensitive to enteric viruses.

Virus infections which may cause severe or life-threatening illnesses meriting chemotherapy are surveyed. Experience has shown that effective antiviral chemotherapy must be given early in the illness, and so this paper concentrates on the earliest symptoms of the various diseases and on methods leading to rapid diagnosis. This can be made by recognizing virus particles of characteristic morphology, if they occur in large numbers. e.g. rotavirus in diarrhoea. Sensitivity may be enhanced by using immune serum to trap viruses. A variety of very sensitive immunological methods is now available for detecting virus-specific proteins in secretions of the respiratory and alimentary tracts. Monoclonal antibodies help to make these more specific. Virus nucleic acids can also be detected in secretions and cells in situ by hybridization with purified radio-labelled nucleic acid. Virus diseases specifically surveyed include varicella zoster and measles, both very important in immuno-suppressed patients; epidemic influenza; and the human papilloma viruses, of which 33 different types are now known, and which cause not common warts on hands and feet, but also venereally transmitted inapparent papillomas upon the surface of the uterine cervix, often leading to cervical carcinoma. The problems of diagnosis of acquired immunodeficiency syndrome are reviewed. The aetiology of chronic non-A, non-B hepatitis is still largely obscure, but a retrovirus may be involved. The diagnosis of rabies is also considered.

The subgroups and serotypes of 178 strains of rotavirus isolated from diarrheic and healthy children in Bangui, Central African Republic, during a 27-month period were determined by enzyme-linked immunosorbent assay. The subgroup was determined for 152 of the viral strains, 18.4% being subgroup I and 81.6% being subgroup II. Of the 143 strains which could be serotyped, 71.3% were serotype 1, 15.4% were serotype 2, and 13.3% were serotype 3. Serotypes 1 and 3 were detected throughout the study, while serotype 2 was detected only during 8 months. No serotype exhibited any special epidemiological properties. The serotypes were found to consist of three different electrophoretypes, two long ones (A and B) and a short one (C). All subgroup I, serotype 2 strains presented short electrophoretypes. Strains with identical long electrophoretypes A were either serotype 1 or serotype 3.

Human, piglet, mouse, foal, lamb, calf and rabbit rotaviruses all infected, but could not readily be subcultured in LLC MK2 cells. Cells infected with mouse and calf rotaviruses reacted by indirect immunofluorescence (FA) with convalescent serum from children, piglets, mice, foals, lambs, calves or rabbits, taken after rotavirus infection. Human, calf, piglet, mouse and foal rotaviruses reacted with human, calf, mouse, foal and lamb convalescent serum by complement fixation (CF). It was not possible to distinguish between different rotaviruses by CF or FA. Neutralization tests, however, detected species-specific rotavirus antigens. Any virus was neutralized by a much higher dilution of homologous species convalescent serum than by any heterologous serum. With the exception of the mouse virus there was very little cross reaction. However, in sera with a very high neutralizing titre for the homologous virus the titre was proportionately raised against heterologous virus. It is, therefore, now possible to type to species an unknown rotavirus by a neutralization test in LLC MK2 cells using convalescent serum from each species.

Summary Faecal filtrates containing rotavirus particles, from children with acute infectious diarrhoea, were inoculated intranasally into gnotobiotic piglets. The piglets developed no symptoms, but virus was readily found by electron microscopy in their faeces during three serial passages. Among 11 piglets tested 3 weeks after inoculation of virus, all had developed fluorescent antibodies against tissue-culture-adapted calf rotavirus but only two had neutralising antibody. Growth of human rotavirus did not occur in either normal or “nude”, thymus-deficient suckling mice.

Summary Extracts of the leaves of Phytolacca americana and partially purified preparations of such extracts caused marked inhibition of the infection of Chenopodium quinoa by cucumber mosaic virus (CMV) and of the infection of monkey kidney cells and embryonated hen eggs by the A2/Hong Kong/1/68 (H3N2) strain of influenza virus. Both viruses became non-infective when mixed with preparations of the extracts but infectivity was regained when the viruses were separated from them by centrifuging.

The Chinese atypical (Group B) rotavirus, previously called the adult diarrhoea rotavirus (ADRV), was examined by transmission electron microscopy using either uranyl acetate or potassium phosphotungstate (PTA) as negative stains. Complete rotavirus particles were seen using uranyl acetate which were indistinguishable morphologically from typical rotaviruses. In the same preparations virus particles with differing degrees of degradation were seen after staining with PTA. This effect was not related to pH of the PTA and could be prevented by fixation of the specimen by 0.1 per cent glutaraldehyde. It is concluded that the use of PTA can give rise to falsely negative results for specimens containing this virus.

A method is described for examining viruses in faeces by direct electron microscopy using negative staining. The particles found in a group of patients with gastroenteritis and a group with other conditions are compared. Small particles in the range of sizes covering parvoviruses and enteroviruses were found about as frequently in each group. Many of these were probably bacteriophages. Many bacteriophages with tails of various sizes and lengths were found. Adenoviruses were found in five of eight patients from whom they were isolated. Reovirus- (or orbivirus-) like particles were found associated with some gastroenteritis patients.

Pleomorphic virus‐like particles have been observed by electron microscopy in the faeces of children and adults with diarrhoea. Some of these particles were approximately 100 nm in diameter and had a “fringe” of closely applied peplomers approximately 10 nm long; they closely resembled Berne virus of horses and Breda virus of calves, the two representatives of a newly proposed family called the Toroviridae. In one sample a toroidal nucleoprotein‐like structure was observed within the particles. For two samples a buoyant density of 1.14 g/ml was determined by centrifugation through a sucrose density gradient. One sample possessed a haemagglutination for rat erythrocytes. The serological relationship between these different viruses was observed by immune electron microscopy, haemagglutination inhibition, and serum neutralisation. The role of these viruslike particles as candidate pathogens of humans is discussed.

A neutralisation test using cell culture and indirect immunofluorescence was applied to isolates of rotavirus from 55 patients with gastroenteritis, in order to determine serotypes. Frequent cross-reactions were observed, but subsequent statistical analysis confirmed the existence of at least three distinct serotypes. Some results, not sufficient for analysis, suggest that at least five serotypes probably exist. It is suggested that two or more viral polypeptides might be involved in neutralisation, one of which might be common to another serotype(s). This would explain the frequent cross-reactions detected by this neutralisation test. Evidence suggesting that some strains may be hybrids is also presented.

Summary Treatment of T3 with various denaturing agents produced distinct components which were identified in the ultracentrifuge. Combining this analysis with electron microscopy made it possible to identify the capsid and nucleocapsid, and a fragment having an s 20, w = 105 ± 10S. This fragment appeared spherical in the electron microscope and had dimensions between 12 nm and 16 nm. It was tentatively identified as the tail. The mol. wt. of the 105S particle and the empty head (200S) were calculated to be 2.3 × 106 and 21.7 × 106, respectively. Adding these weights to the estimated weight for DNA (25 × 106, Lang & Coates, 1968) gave a total weight for the phage of 49 × 106, equal to the mol. wt. obtained by Swaby (1959). In addition to these large fragments, two peptides were examined, one had a mol. wt. of 9300 ± 1400 in 6 m-GuHCl (1.5S) and appeared to be a single chain; the other, released when a suspension of phage was diluted, had a mol. wt. of less than 10000. The origins of the two peptides remain largely speculative, but in view of the marked associative properties of the 1.5S peptide and its detection only when the head was destroyed, it is likely that it was a binding fraction in the head of the phage. The second peptide may be more intimately associated with the DNA of T3.

SUMMARY: The changes in fowl-plague infected tissue cultures have been investigated by the phase-contrast, and electron microscopes. The nucleolus of infected cells first increased in size and then became granular and highly refractile, finally disintegrating into fragments about 64 hr. after infection. After 36– 48 hr., when the inclusions were already developed, filaments were found, usually in bundles, possibly having escaped from the nucleus into the cytoplasm; a few filaments showed segmentation into spherical bodies and shorter filaments. A possible life cycle is suggested for the virus: first, penetration of the nucleus and growth within it, then extension into the cytoplasm in a filamentous form, and eventual segmentation of some of the filaments into elementary bodies.

A case is described of herpetic hepatitis in a pregnant woman with primary herpetic stomatitis. Intranuclear inclusion bodies and virus particles were found in hepatocytes, and herpes virus was isolated from a liver biopsy and from oral swabs but not from blood. From rising titres of neutralizing and complement-fixing antibody it is concluded that the oral infection was a primary one. Factors predisposing to the hepatitis are discussed.

A fatal case of vaccinia necrosum treated with antivaccinial gamma globulin and N-methyl isatin beta-thiosemicarbazone (33T57) is described. Histological abnormalities found at necropsy included intranuclear as well as cytoplasmic inclusion bodies, absence of lymphoid germinal centres, grey hepatization, and bronchiolar epithelial hyperplasia. Virus titres were highest in the original skin lesion; virus was also found in the lymph node draining it and in the kidney and the brain. No toxic effects could be clearly attributed to the drug used in treatment.

Summary The reticular structure previously described on the original taylor 1233 strain of influenza C was found on two further strains of influenza C. In negatively stained preparations of influenza C three kinds of particles with different surface morphology have been found: particles with ‘spikes’ typical of myxovirus, particles with a hexagonal reticular surface and ‘naked’ particles with no reticular structure. The fine structures of influenza viruses A and C as seen in thin sections were compared. In walls of influenza A the structures encountered from the surface inwards were: a layer of spikes hollow and open at both ends; an electron-transparent layer of uniform width; and a finely granulated layer. The electron-transparent layer of influenza C virus was much less prominent; the spikes were irregularly placed and often absent. On sectioned influenza C filaments a hexagonal pattern was seen. The internal component of influenza A virus measured about 6 nm. in diameter and sometimes appeared to be hollow. The internal component of influenza C measured about 9 nm. across and appeared to be helical. On the basis of these results the taxonomic position of influenza C has been questioned.

The reovirus-like particles present in faeces of young children with acute gastroenteritis and the virus causing acute diarrhœa in newborn calves were found to be indistinguishable from each other in size and shape. The human and calf viruses were agglutinated by sera from convalescent calves. Some adult human sera neutralised infectivity of the calf virus in tissue-cultures. Sera from convalescent children agglutinated both human and calf viruses and gave a positive immunofluorescent reaction with calf virus in cell-cultures. Since these viruses differ morphologically both from reoviruses and orbiviruses, the name " rotavirus " is suggested for them.

Pleomorphic virus-like particles about 100 nm in diameter with a fringe of closely applied peplomers (7-9 nm in length) were observed by electron microscopy in the stools of 20 children and adults with gastroenteritis. In most of the samples no other viral or bacterial pathogens were detected. In form and under immune electron microscopy these virus-like particles resembled the Breda virus isolated from diarrhoeic calves. These objects may be a viral pathogen of humans.

During 1974 in the West Midlands of England, 38% of children less than 6 years old with enteritis were excreting rotaviruses. Children aged from 6 months to 3 years were those most commonly infected. Rotavirus infections were most common during winter with only a few sporadic cases during summer. A possible pathogen was detected in 59% of patients examined.

The reovirus-like particles present in the feces of young pigs and foals with acute enteritis and the virus causing epizootic diarrhea of infant mice were found to be indistinguishable morphologically from each other, from the South African SA. 11 and "O" viruses, and from the rotaviruses of children and calves. The inner capsid layer of each of these viruses reacted seriologically with sera of children, calves, mice, piglets, and foals convalescent from infection with their respective rotaviruses. These sera reacted by immunofluorescence with human, bovine, porcine, and murine rotaviruses, SA.11, and "O" viruses in tissue cultures and with human bovine, procine, nad murine viral antigens by complement fixation and gel diffusion. However, the antisera differed in their ability to react serologically with the outer capsid layer of the viruses investigated and in their ability to neutralize tissue culture-adapted calf virus. These two tests may demonstrate strain or host specificity among rotaviruses. Since the porcine, murine, and equine viruses are closely related serologically to and are morphologically identical to the human and bovine viruses, they should be included in the group of viruses for which the term "rotavirus" has been suggested. All known members of this proposed group of viruses share a common antigen, probably situated within the inner capsid layer; thus, any one of the viruses may be used for the preparation of antigen or antibody for diagnostic tests, and this will aid in the diagnosis of virus infection in those species from which a rotavirus has not been cultured.

Examination of the intestinal contents of free-living Oryzomys nigripes rats by PAGE revealed two sharply defined bands that could be stained by ethidium bromide or by silver nitrate with comparable intensities. The molecules forming these bands were susceptible to digestion by pancreatic RNase A but not by RNase T1 or by DNase I. Their lengths were estimated to be about 2.6 and 1.5 kbp, respectively, by comparison with rotavirus SA11 genome segments. They cosedimented in CsCl gradients at a density of 1.39 to 1.40 g/ml, together with uniform particles approximately 35 nm in diameter with indistinct surface structure. It is suggested that these particles represent an as yet undescribed virus with a bisegmented double-stranded RNA genome, for which the name 'picobirnavirus' is proposed.

A method is described for the serotyping and subgrouping of rotaviruses by enzyme-linked immuno-sorbent assay (ELISA). For the isolates tested “blind” under code in parallel the serotyping results obtained by ELISA and serum neutralization of fluorescent focus formation were the same. After absorption of the typing antisera with purified heterotypic rotavirus, up to 128-fold differences in titres between isolates were observed in the ELISA test. The results of serotyping and subgrouping, by ELISA, of strains previously described and characterized by the authors' and other laboratories are also given. An attempt is made to correlate the serological findings with variations between virus strains based upon patterns of genome molecular weights as revealed by polyacrylamide gel electrophoresis. The application of this ELISA technique to epidemiological studies and vaccine research is discussed.

Enzyme immunoassays were developed for the detection of Breda virus antibody and antigen. Cattle sera collected in the United Kingdom were found to have a high prevalence of antibody (55%) to Breda virus when examined in a competitive enzyme-linked immunosorbent assay. A low prevalence of antibody was found in pigs (2.2%), and no antibody was found in sheep or goat sera. No antibody to either Breda virus or Berne virus was detected in human sera collected from veterinarians and farm workers. Only 1 of 430 human fecal specimens (0.2%) contained Breda virus antigen detectable by enzyme-linked immunosorbent assay.

Virus particles resembling reoviruses or orbiviruses were found in the faeces of 40 of 73 patients under 6 years of age with acute gastroenteritis and in faeces of only two babies among 31 patients under 6 years admitted to hospital with other diagnoses. In morphology the particles resemble orbiviruses more closely than reoviruses, but differ in appearance from the orbiviruses in having a smooth, circular outline with a well marked continuous rim as seen in negatively stained preparations. They appear not to be serologically related to reovirus types 1, 2, or 3 and may be members of a new group.

The influence of divalent cations on the stability of human rotavirus was investigated using the indirect immunofluorescence (FA) technique in LLC-MK2 cells to titrate infectivity. Rotavirus infectivity was stabilized by calcium and strontium ions in solution but not by magnesium ions. Rotavirus isolates were found to be much less stable at 37 degrees C than at + 4 degrees C or 20 degrees. A study of virus morphology at intervals during the course of the experiment and treatment with the chelating agents EDTA and EGTA suggests that loss of infectivity coincides with the removal of the outer capsid layer and that calcium may be required to maintain virus integrity.

Between 1983 and 1988, subgroups and serotypes were determined for 907 of 1,084 clinical specimens of rotaviruses collected in various countries of Europe, North and South America, Africa, and Asia. Enhanced enzyme immunoassays based on monoclonal antibodies specific for rotavirus proteins VP6 and VP7 were used. Significant differences in the prevalent serotypes were detected from year to year in the United Kingdom and Brazil and also in different countries during the same year. Throughout the study, rotavirus serotype 1 was detected most often (53.8%), followed in frequency by serotype 2 (17.8%), serotype 3 (12.1%), serotype 4 (11.1%), and serotypes other than 1 to 4 (5.1%). No individual serotype was found to predominate consistently in any one location. In the United Kingdom, rotavirus serotypes varied in prevalence in a regular but not predictable way. We suggest that a similar epidemiology might be found in other settings. Seventeen unusual strains were detected. Of these, five strains did not react with reference monoclonal antibodies specific for subgroup I and subgroup II, but they reacted with rotavirus group A-specific polyclonal and monoclonal antibodies; four strains were of subgroup II, serotype 2, and at least one had a "long" electropherotype; two strains were of subgroup I, serotype 2 with a long electropherotype; and one strain was of subgroup I, serotype 3. Five group C rotaviruses were detected.

Several viruses causing infections of the gastrointestinal tract, unknown until recent years, can be detected rapidly by direct electron microscopy of faeces from patients (and from some animals also). A rapid tissue culture technique is available; its most important application is for rotaviruses. Immune electron microscopy has been of great service in detecting some of these viruses and in elucidating their serological identities.

Summary Rotavirus subgroups and serotypes were determined in 61 rotavirus-positive faecal samples obtained from children living in Belem, Brazil, followed up from birth to 3 years of age. Fifty-five (90%) of the specimens were subgrouped and the serotypes of 30 (49%) of them were determined. Subgroup II was detected in 49 (89%) of the 55 subgrouped strains. Serotype 1 was present in 15 (50%) of, the 30 serotyped samples; serotypes 2, 3 and 4 were found in 30%, 3.3% and 16.7% respectively, of these specimens. Absence of Vp7, the major outer capsid glycoprotein, did not allow serotyping in 21 (34.4%) of the 61 rotavirus-positive specimens, and an unidentifiable new serotype was found in faeces of one child. In addition, 4 samples were classified as subgroup II serotype 2 (which is very unusual). Twelve (80%) of the 15 serotype 1 (subgroup II) specimens were collected from children (5 of them asymptomatic) during their first year of life. All 9 serotype 2 (subgroups I, II, or not determined) samples were detected during the second and third years of life, 7 (77.8%) of them were related to apparent infections. The 5 serotype 4 (subgroup II) samples were obtained throughout the study period, and were associated with both symptomatic (3 cases) and asymptomatic infections. Thirteen children had more than 1 rotavirus infection. Three had 3 successive infections. In 3 cases, the initial infection (either symptomatic or asymptomatic) caused by serotype 1, was followed by a subsequent diarrhoeic episode associated with serotype 2.

SUMMARYInfluenza C virus antigen could be detected by immunofluorescence in chick amniotic epithelium 4 hr. after inoculation with undiluted amniotic fluid. Antigen could be detected in the nucleus of rhesus monkey kidney tissue cultures 10 hr. after infection. Only cytoplasmic fluorescence was observed 17–21 hr. after infection.The pattern of replication was similar to that reported for group I myxoviruses, indicating that the virus is correctly classified as an influenza virus.

Seven cases are described of highly infectious disease characterized by vesicular eruptions in the mouth and on the hands and feet. In previous outbreaks described in the literature Coxsackie type A16 virus was isolated but from these cases, similar clinically, Coxsackie type A5 virus was isolated.

Sections of chorio-allantoic membranes fixed at intervals after infection with large doses of vaccinia virus showed almost homogeneous areas appearing in the cytoplasm after 4 hr.; these areas contained immature forms of virus which had a thin surface membrane and an eccentric ‘nucleoid’, and were surrounded by mitochondria. Mature forms of virus, having a larger central nucleus-like structure and a thicker cortex, were first found in areas of virus growth 10 hr. after inoculation. Similar immature and mature forms were found in cells infected with fowl-pox, myxomatosis and ectromelia viruses. It is concluded that the immature forms are of low infectivity or not infective.

Enzyme-linked immunosorbent assays were developed for the detection of group B rotavirus antigen and antibody. The specificities of both assays were evaluated for antigens and serum specific for rotavirus groups A to D. Serum collected in the United Kingdom from different animal species exhibited the following high prevalence of group B rotavirus-specific antibody: pigs, 97%; cattle, 71%; sheep, 91%; and goats, 91%. In human serum, a lower prevalence of group B-specific antibody was detected; serum from blood donors showed 10% prevalence, and serum from veterinarians showed 4% prevalence.

Two outbreaks of acute gastroenteritis occurred in 1974 in a long-stay children's ward. Electron microscopy demonstrated rotaviruses in faeces from the affected children in the first outbreak, and adenoviruses in faeces from affected children and a nurse in the second outbreak. The illness in both outbreaks was very mild; but the diarrhœa associated with rotavirus infection usually lasted 5-8 days (in one patient it lasted for 28 days) and sometimes started with vomiting; whereas the adenovirus-associated diarrhœa lasted only 2-4 days and was not associated with vomiting. Neither the rotaviruses nor the adenoviruses could be established in tissue-culture.

Research on plant virus vectors has shown that almost all types of organisms feeding on or parasitizing infected plants, including sucking and biting insects, mites, nematodes and chytrid fungi, can act as specific vectors. Surprisingly, however, natural vectors have not been found for some of the most infectious of the plant viruses such as tomato bushy stunt virus (TBSV) and tobacco mosaic virus (TMV). Here we report that when purified TBSV was consumed by human volunteers, it subsequently occurred in their faeces and could be detected by mechanical inoculation of faecal extracts to the leaves of an indicator plant (Chenopodium quinoa) in which the virus causes countable, discrete lesions. Thus, the virus remained stable on passage through the alimentary tract. The significance of man as a carrier, able to disseminate the virus widely within the environment, is discussed in relation to the epidemiology of the disease. To our knowledge, this is the first demonstration of an ‘alimentary-resistant’ plant virus.

THE inhibition of haemagglutination by indicator influenza virus using mucoprotein inhibitors is a well-established observation1. It has been proposed that for a molecule to inhibit viral haemagglutination it must not only contain sialic acid as an end-group susceptible to neuraminidase, but also be of ‘minimum size’2. Orosomucoid3, a serum glycoprotein containing 11 per cent of neuraminidase-susceptible sialic acid, however, does not inhibit viral haemagglutination4. It has been suggested that orosomucoid may fail as an inhibitor because its molecular weight of 41,000 is too low2. We have found that incubation of this glycoprotoin at 80° C caused it to polymerize in solution and to become an active inhibitor of certain strains of influenza virus5. Further observations on this phenomenon are reported here.

An epidemic of acute hepatitis B followed the administration of human immunoglobulin to members of the staff of a mission hospital in India and their families. Jaundice developed in 123 (38%) of 325 persons inoculated. Hepatitis-B surface antigen was detected in three of the batches of immunoglobulin which were available for testing.