Introduction

The dawn of virology, encompassing the years between the end of the nineteenth century and first decades of the twentieth, was to some extent erratic, even awkward. This was due to a variety of reasons including the following: (i) the bacterial world was on the spotlight at that time, and all diseases were assumed to be caused by bacteria; (ii) the opinion of medical practitioners ordinarily prevailed over that of microbiologists, as Microbiology was considered a “second-class science.” Hence, often, the priority was to find a remedy rather than the cause of the disease; (iii) the relevant technology was still missing, since most viruses, with the exception of some poxviridae, could only be observed under the electron microscope; and (iv) both microbiologists and medical practitioners often disregarded one of the main viral characteristics, namely their filterability and their inability to grow independently, in the absence of host cells.

The present mini-review describes misconceptions in several viral diseases that caused considerable human suffering and are responsible for multitude of pandemics throughout our history. The diseases were selected to include both illnesses correctly identified as of viral ology and those that, for whatever reason, were misidentified as caused by bacteria.

Influenza

A simple perusal at the table of contents of the 1921 influenza publication by Twort and Twort would bemuse current readers (both microbiologists and lay people), in particular, statements such as those found in chapters II: “The bacteriology of Influenza”; III: “Conditions influencing the toxicity and pathogenicity of the influenza bacillus”; or IX: “The cultivation of the influenza bacillus in symbiosis with amoebae on blood agar, etc.”

The article, however, must be read carefully and in the context of its time, when there was considerable pressure to elucidate the origin of influenza. It was published just 1 year after the commonly named “Spanish flu” epidemic (January 1918–December 1920), during the second wave of the pandemic involving the N1H1 orthomyxovirus. This pandemic infected 500 million people all over the world, including remote areas, and resulted in 50 to 100 million casualties (3 to 6% of the human population at the time). The name “Spanish flu” arose because Spain’s neutrality during the Great War allowed its newspapers to freely report influenza cases and deaths, while wartime media censorship in the countries involved in World War I prevented journalists from reporting them. This created the false impression that Spain was the origin of influenza pandemic.

Twort and Twort (1921) conclude “that influenza is caused by B. influenzae.....No new type of bacterium was discovered, and no evidence was obtained of the presence of an ultra-microscopic virus …..,” and “the influenza bacillus grows in symbiosis with amoebae on blood agar”……“the influenza bacillus grows in symbiosis with a small spirillum that was isolated from a grass emulsion…..” The authors’ conclusion that a bacterium (previously known as Pfeiffer’s bacillus; Pfeiffer 1893) causes influenza could be due to social pressure. The morbidity and mortality of the disease made it urgent to elucidate the origin of influenza, while it was almost generally assumed that influenza was caused by Bacillus influenzae, as described by Wollstein and Meltzer (1912) (these authors were among the first to report a Streptococcus as the leading cause of pneumonia worldwide).

However, this misconception is hard to comprehend, as Frederick W. Twort was the person who in 1915 discovered bacteriophages, ultra-filterable organisms similar to the influenza virus (Duckworth 1976). Indeed, the filterability of the influenza etiological agent was reported by Nicole and Lebailly (1918) and by Gibson et al., 1918. Furthermore, Olitsky and Gates (1921) described that “the essential effects (influenza) were produced by a substance wholly unrelated to these bacteria” (the bacteria normally isolated from the throats of patients suffering from influenza), and the authors concluded that the cause of influenza was a multiplying agent and not just a passive substance. Nevertheless, there was a widespread belief at the time, as per Hort’s hypothesis, that Pfeiffer’s bacillus (later classified as Haemophilus influenzae) would sporadically undergo a filterable stage, as well as occasionally exhibiting fungal developmental growth (Wade and Manalang 1920). It was also believed that a majority of, if not all, known bacteria could exist in a filterable and in a non-filterable state (this idea was proposed and maintained by Arthur Isaac Kendall as late as 1931 (Kendall 1931)).

As per usual, vaccination was always proposed as the best way to control influenza epidemics (see Heagerty 1919). Once more, this would have increased the social pressure to isolate the causative agent, in order to prepare the antigens to use in vaccination. The efficiency of this vaccination was demonstrated by Haythorn (University of Pittsburgh) in 1920 (Haythorn 1920): “Among the unvaccinated the case incidence was 9.8 per cent, and deaths were 1.2 per cent; among the vaccinated there were no deaths, and 1.2 per cent only was attacked.”

The origin of influenza was much debated in the years following the Spanish flu pandemic, until Smith, Andrewes, and Laidlaw (1933) finally identified the causative agent as a virus. The true viral nature of influenza’s causative agent was confirmed when Siemens released the first commercial electron microscope in 1938, thus settling the controversy. In reality, this historic misunderstanding concerning influenza stems from the original publication by Pfeiffer (1893), a German physician and bacteriologist. He was so highly regarded scientifically that his opinion that influenza was caused by a bacterium (Bacillus influenza) was generally accepted.

A group of bacteria, intracellular parasites belonging to the family Rickettsiaceae (for instance, the causative agent of Q fever), were also discovered at this time but, although filterable, they were always identified as bacteria and not as viruses. The etiological agent of Q fever was originally named Rickettsia diaporica by Derrick, Burnet, Cox, and Davis, because of its ability to pass through filter pores, just as viruses do (McDade 1990). This parasite was renamed as R. burnetti (Derrick 1944) and finally Coxiella burnetti in 1948. Although uncommon, the bacterium can be found in a variety mammals (including dogs, cats, and cattle) and invades human beings through the respiratory tract (rarely by ticks, as opposed to other rickettsial diseases). C. burnetti is currently classified as a gammaproteobacteria related to Legionella sp. In infected humans, the parasite develops intracellularly and causes a syndrome known as Q fever (Q stands for “query” because the cause of the disease was not known), a name coined by Derrick (1935), when studying the causative agent in Brisbane (Queensland, Australia; Derrick 1944; see Fig. 1 for all considerations concerning this mini-review).

Fig. 1
figure 1

Some hits and misinterpretations in the history of early virology. True (a) and erroneous (b) findings. References in order of appearance in the chronological line: (Heine 1840), (Herrick 1878), (Chamberland 1884), (Pasteur 1885), (Finlay 1903), (Gordon 1937), (Canon and Pielicke 1892), (Ivanovski 1892), (Pfeiffer 1893), (Sanarelli 1897), (Beijerinck 1898), (Bury 1902), (Landsteiner and Popper 1909), (Duckworth 1976), (Twort and Twort 1921), (Ledingham 1931), (Knoll and Ruska 1932), (Smith et al. 1933), (Stanley 1935), (Kausche et al. 1939), (Reagan and Brueckner 1952) 

Full size image

Rabies

Rabies is a viral disease that produces central nervous system inflammation in mammals, including humans. This disease is almost always fatal once symptoms appear. Hampered by the elusive nature of the causal agent, as is usually the case for viral diseases, isolation and identification of the rabies Lyssavirus had to wait until the electron microscope became available. Louis Pasteur, despite this handicap, was able to isolate the rabies virus and create an attenuated strain that worked as an effective vaccine against the disease.

Long before Louis Pasteur, Dr. Percival (1789) published an article on the nature of canine rabies stating “There seems to be a striking resemblance, in many particulars, between some species of tetanus and the rabies canine” and adding “The virus in this case, if virus is to be supposed………” However, he was reluctant to ascribe hydrophobia to a tetanus manifestation, although he was influenced by Dr. Rush of Philadelphia that in the same year concluded “The more I have considered the causes and symptoms of hydrophobia, the more I am disposed to ascribe it to the same proximate cause as the tetanus” (Rush 1794). In 1817, Webster (Webster 1817) published the conclusion that a sort of poison was introduced into the human body when bitten by a rabid animal, but our understanding is that he did not elaborate on the nature of the etiological agent of rabies. Notwithstanding, medical practitioners of that period usually treated the bite with quicksilver (or by surgical excision, if possible) as a way to control the spreading of the causative agent. Post-mortem examinations of the rabies victims, if carried out immediately after death, always revealed vast inflammation of the whole central nervous system (including the spinal cord) and considerable water accumulation under the tunica arachnoides and cortical convolutions. The idea of a poisonous substance transmitted by the bite of a rabies-infected animal to its victim also appears in the article published by Johnson (1819), but again, he provided no insight into the nature of the pathological agent. Rabies treatment involved, apart from quicksilver, the application of natural herbs including the lichen Cinereus terrestris. Prince (1872), following a prescription from George Dampier (1625), reported that this treatment was successful if applied immediately after the bite.

In addition to the discoveries described above, it was the tremendous work carried out by Louis Pasteur that finally produced a vaccine for rabies. Despite his vast knowledge on bacteriology and other microorganisms, including yeasts, Pasteur was unable to detect a bacterium in active cases of rabies. Nevertheless, he isolated the etiological agent and used it to produce a successful vaccine against the disease; in fact, Pasteur’s strain is still currently in use. Micrococci were often observed in samples taken from rabies patients, but these organisms were never isolated in culture and were shown not to be the pathogenic cause of rabies (Horsley 1889). Hence, in the case of rabies, as opposed to influenza, bacteria were never considered the etiological agent of the disease. Pasteur’s (Pasteur 1885) conclusions on rabies were tested by multiple studies carried out between 1885 and 1890, but all confirmed Pasteur’s results; these publications include Vignal (1886), Horsley (1889), Adami (1889), and Ernst (1891).

Poliomyelitis

Human poliomyelitis, a disease that affects the central nervous system, is caused by an enterovirus belonging to the family Picornaviridae. (Please note that chronic syphilitic poliomyelitis, caused by Treponema pallidum, that also affects the central nervous system will not be dealt with here.) Although the history of poliomyelitis infections extends into prehistory, the first medical report on the disease occurred in 1840 and poliomyelitis was then recognized as a defined disease (Heine 1840). Sturge (1879) indicates that in 1879 and 1880, the disease became known as “spinal infantile paralysis”; he also states that the morbid changes (acute inflammation) in the patient were principally confined to the anterior horns of gray matter. But neither Sturge, nor any earlier authors, elaborated on the nature of the etiological agent of poliomyelitis, although it is known that some practitioners unsuccessfully searched for Koch’s bacillus in patients and treated subacute forms of the disease with compounds such as ergotin and iodoform (Smith 1883). In fact, microorganisms, usually diplococci, were found in either the cerebrospinal fluid or in the anterior horns of patients (Wollstein 1908). Indeed, either microorganisms or their toxins were considered the cause of the disease, as many of the symptoms (such as spinal lesions) of poliomyelitis could be reproduced by injecting animals with either Corynebacterium diphtheriae, Streptococcus pyogenes group A β-hemolytic, or even Escherichia coli (Bury 1902). In 1908, Martha Wolstein published a considerable body of work on the etiology of poliomyelitis; this publication did not support the conclusion of a bacterial origin for the disease. Remarkably, as opposed to influenza, to the best of our knowledge, there is no publication suggesting bacteria as the etiological cause of poliomyelitis. The causative agent was identified as a virus (smaller size than bacteria and hence filterable) by Landsteiner and Popper (1909), although confirmation and visualization of the viral particles was finally achieved by electron microscopy.

Smallpox

Smallpox has been one of the greatest killers throughout human history (particularly lethal for the isolated non-occidental populations, such as those inhabiting the New World or Siberia) until global eradication in 1980, due to vaccination with live vaccinia virus. The orthopoxvirus (variola) etiological agent of the disease appears to have evolved from an African rodent virus between 68,000 and 16,000 years ago (Esposito et al. 2006)

Although Jenner (1786) is considered the key researcher in the vaccine development against smallpox, some 20 years before, the farmer Benjamin Jesty successfully protected his family from smallpox by inoculating them with cowpox. Indeed, Jesty was not even the first to achieve protection against smallpox, as Bose (1769) made the same discovery in Germany in 1769. Furthermore, smallpox prevention was reported at an earlier time in both China and India (Thèves et al. 2014). Undoubtedly, these early treatments paved the way for Jenner to develop his ideas on vaccination against smallpox, although those early publications did not discuss the biological nature of the causative agent of smallpox.

The typical inclusion bodies (representing sites of viral multiplication) found in epithelial cells infected with smallpox (known as Güarnieri bodies) were first thought as living beings and assigned as the cause of the disease. They were reported as “elementary bodies” by authors like Buist (1886) (see Gordon 1937 and Calmette and Guérin, 1901), or “spores” by Councilman et al. (1903). But these researchers did not provide any additional information on the nature of the etiological agent. However, the particles or “spores” causing smallpox were shown to be filterable and fully active. As the smallpox virus is the largest animal virus, as opposed to other viruses, it is possible to visualize it using the light microscope. This helped Ledingham identify the elementary bodies as the true cause of smallpox in 1931 (Ledingham 1931). The size of the virus, coupled with the use of electron microscopy, resulted in the early identification of the chemical nature of the smallpox virus, long before other viruses were characterized. This also contributed to the fact that the virus was never mistaken for a bacterium, as happened with influenza (Fenner et al. 1988).

Yellow fever

Yellow fever (also known as Bulan fever or black vomit) became well known and was frequently described by both Spanish and British colonizers of tropical and semi-tropical regions in the new world. Many period writings include descriptions of the disease, including diagnosis and ways of transmission, but nothing is found on the nature of the causative agent. However, it was known that the illness could be transmitted from person to person (although this rarely happened) and that the disease originated from “marsh miasmata” (Fergusson 1817). This was particularly evident to Mortimer, 1817, published his 10 years of experience and studies on Bulan fever. At that early stage in the understanding of the disease, the researchers managed to measure the distance traveled by the causative agent of the black vomit, 40 ft per day (White 1876). Herrick (1878) assumed that yellow fever was produced by a “germ” that reproduced outside the human body but, to the best of our knowledge, he did not mention either the nature of the “germ” or the route of acquisition of the disease. Indeed, he was totally ignorant on the role of a female mosquito, Aedes aegypti or Stegomyia fasciata, as a vector in the transmission yellow fever. Both medical practitioners and surgeons realized, however, that this fever was somewhat different to the classic fevers known at that time, since it only occurred in certain areas, seasons, and latitudes, with the peculiarity that it mostly occurred in the cities, as opposed to rural areas (Sternberg 1879). In towns such as in Memphis, New Orleans, and Baton Rouge (USA), the measures to contain and prevent yellow fever included periodically emptying and cleaning vaults and cisterns. But, most importantly, all water reservoirs were filled to a minimum depth of 3 ft. The later proved essential, as female mosquitoes need shallow water to lay eggs; hence, this practice hindered mosquito reproduction (Tyner 1879). Charles Finlay was one of the first to recognize the involvement of mosquitoes, while working in Havana, he mentioned in a hospital report in 1886 (published in 1903) (Finlay 1903): “In the light of modern research yellow fever must now be considered as a disease produced, in individuals belonging to certain human races, by a two-host, microscopic, perhaps ultra-microscopic germ, endo-parasite of the female of a particular species of mosquito-the Stegomyia fasciata (Theobald).” He asserted that a small protozoon, referred to as Myxococcidium stegomyiae by some authors (see McL (1904)), could be the etiological agent.

Finlay’s publication was the first indication that the causative agent of yellow fever was not a bacterium. In fact, the only dissenting report came from Giuseppe Sanarelli (Sanarelli 1897), an Italian bacteriologist working in Uruguay, who in 1897 mistakenly reported that yellow fever was caused by a bacterium he named Bacillus icteroides.

Measles

Munro (1891) described measles as “the opprobrium of the sanitary Science in the 20th century, without control by preventive medicine.” Measles is a highly contagious infectious disease caused by the measles virus and characterized by coryza, Koplik’s spots, and a skin rash. The measles virus is an enveloped (non-segmented) RNA virus belonging to the genus Morbilivirus (family Paramyxoviridae) believed to have emerged as a zoonotic disease from rinderpest (an infectious disease of cattle) sometime between 1100 and 1200 AD (Furuse et al. 2010). The virus was first isolated in 1954 by Enders and Peebles (Enders and Peebles 1954). Although measles is a preventable disease through vaccination [the most successful vaccine against measles was developed by Hilleman et al. (1968)], it annually affects around 20 million people worldwide and causes many deaths (in excess of 70,000 in 2014).

A 1892 publication summarizing the work of several practitioners (Canon and Pielicke (1892) among others) claimed in 1892 that measles was caused by “one and the same bacillus.” The disease was occasionally mistaken for smallpox, even until the first decade of the twentieth century, although a differential diagnosis was well known to Arab physicians (Butler 1913). Except from the publication mentioned above on the possible bacterial nature of the causative agent of measles, other reports by both microbiologists and medical practitioners were reluctant to unequivocally identify a bacterium as the etiological agent of the disease, as opposed to what happened with influenza or yellow fever. An example of this ambiguity is the following paragraph from Sellards and Bigelow (1921): “Since measles is an acute self-limited disease followed by a well-marked degree of immunity, it could hardly be caused by a typical protozoon. According to the usual views, it is a droplet born infection and its portal of entry is by way of the mucous membrane of the upper respiratory tract. These considerations suggest an aerobe or facultative anaerobe rather than an obligatory anaerobic organism.” The authors adopted a technique, popular at the time to grow bacteria, which involved the use of chocolate agar as used in the cultivation of the influenza bacillus. Although the causative agent was not specifically described, up to 1921, the microorganism was treated as a normal bacterium, with the exception that it was filtratable, as opposed to the majority of bacteria. Duval and D’Aunoy (1922) were the first authors to suggest the word “virus” to refer to the causative agent of measles. Nevertheless, later publications still claimed a bacterial origin for the disease. In particular, Ferry (1927) reported isolating Streptococcus viridans from the blood of measles patients, which he named S. morbilli, and 2 years later, Smith reported the isolation of a green coccus (Smith 1929). These studies, as well as others not cited here, were clearly influenced by the work of Hektoen (1918) and Thomson (1923) and created confusion on the etiological origin of measles.

German measles

German measles, also known as rötheln, rubella, or 3-day measles, is an infection produced by a virus (Rubivirus) that belongs to the Togaviridae family and also causes the congenital rubella syndrome. The microbial origin of this disease was obfuscated during the eighteenth, nineteenth, and first decades of twentieth century, and medical practitioners often misdiagnosed this as measles or smallpox, or even as a form of scarlet fever. The term “rubella” (from the Latin word, meaning “little red”) was coined by the German physician Hildenbrand (see Foss 1872), although previous observations made by De Berger in 1752 clearly differentiated the exanthematous disease from both scarlet fever and measles (Corbin 1908). Singularly among viral diseases, no bacterium was ever considered as the etiologic agent of Rubella; in fact, from the start, it was generally agreed that the disease was caused by a filterable virus and that its mode of transmission was very similar to that of measles and chickenpox.

The evidence presented here highlights the fact that deciphering novel diseases requires well-prepared scientists that strictly follow the empirical “scientific method” while keeping an open mind to elucidate the new findings and to present them to the society in an appropriate way. The six paradigms presented above are past specific examples of how research can go wrong when conjectures are accepted as facts or when there is dissent, instead of collaboration, between medical practitioners and microbiologists. The errors described above originated from medical practitioners and microbiologists that were reluctant to scrutinize and challenge the ological origin of a disease accepted at the time, without considering the emerging scientific knowledge and new discoveries. Hence, at that time, researchers accepted without question the viral origin of a disease that was in fact caused by a bacterium, while advances in bacteriology resulted in many researchers assuming a bacterial origin to many diseases, even those caused by viruses.