THE YALE JOURNAL OF BIOLOGY AND MEDICINE 53 (1980), 19-25

Virus Morphology as an Aid for Rapid Diagnosis

FRANCES W. DOANE

Department of Microbiology and Parasitology, Faculty of Medicine,University of Toronto, Toronto, Ontario, Canada

Received August 9, 1979

Standard methods of virus diagnosis may take many days to complete. As antiviral drugs arebeing used with more effectiveness, it becomes more important to develop rapid diagnosticmethods. It takes only a few minutes to prepare and examine a specimen for electronmicroscopy (EM), using the negative staining technique. Viruses in the specimen can readily beidentified by their morphology. In order to be detected by EM there must be at least 107 virusparticles per milliliter of sample. This concentration is frequently found in certain types ofspecimens. The sensitivity of EM is increased 100-fold if homologous antibody is used toaggregate the virus. Visualization of virus-antibody aggregates forms the basis for serotyping byimmunoelectron microscopy (IEM).

The commonly practised laboratory methods to detect the presence of a virus inclinical material have changed little in the past 25 years. A specimen is inoculated intoa host system (tissue culture, eggs, animals), and subsequent detection and identifica-tion of an isolated virus depends on indicators such as cytopathic effect, hemagglu-tinin, complement fixation, etc. Results obtained by these procedures usually takeseveral days-if not weeks-to complete. Thus, although they continue to serve asthe backbone of most virus laboratories, current procedures are rarely capable ofproviding a rapid diagnosis.With the advent of specific antiviral drugs, especially for herpesviruses, there is

now more pressure on the diagnostic virologist to provide results quickly. Directmicroscopic examination of the clinical specimen for the virus itself is one obviousapproach. For all practical purposes, viruses are too small to be seen by lightmicroscopy, but they can easily be visualized, and morphologically identified, byelectron microscopy (EM).The detection and identification of vruses by EM offers many advantages. Most

important-the method is fast; total time required for specimen preparation and EMexamination is rarely more than 30 minutes. Requirement for an isolation system isremoved, and loss of infectivity of virus in the specimen becomes unimportant. Thusviruses that are difficult or impossible to culture can be identified by EM. Family orgroup identification can be made immediately, entirely on the basis of virusmorphology. And by immunoelectron microscopy-serotyping directly on the EMspecimen grid-type-specific identification of a virus can often be accomplished.The following brief review will examine some of the ways in which virus morphol-

ogy can serve as an important aid for rapid virus diagnosis, some of the limitations ofthis approach, and current and future developments in electron microscopy relating

19

Address reprint requests to: F.W. Doane, M.A., Department of Microbiology and Parasitology,Faculty of Medicine, University of Toronto, Toronto, Ontario, M5S IAl, CanadaCopyright i 1980 by The Yale Journal of Biology and Medicine, Inc.All rights of reproduction in any form reserved.

FRANCES W. DOANE

.BPLATE 1. Viruses seen by electron microscopy in negatively stained vesicle fluid. x 134,000. A.

Herpes virus from chicken pox. Note icosahedral nucleocapsid surrounded by envelope. B. Vacciniavirus, exhibiting the brick shape and thread-like surface characteristic of poxviruses. The black materialseen in these two micrographs is the contrast-enhancing negative stain.

Ap i.

kx" ~PLATE 2. Elec-tron micrograph ofnegatively stainedparamyxoviruses frominfected cell culture.These viruses are ex-

tremely pleomorphic,but are easily distin-

p7 guished by their fine-fringed envelope sur-

rounding masses ofcoiled nucleocapsids.

PLATE 3. An ex-

ample of the virus-antibody aggregateseen in immunoelec-tron microscopy. Thevirus is echo 11 fromcell culture. x 151,000.

20

VIRUS MORPHOLOGY IN RAPID DIAGNOSIS

to diagnostic virology. For detailed information on electron microscopy in diagnosticvirology, the reader is referred to more comprehensive reviews [1,2].

DIRECT EM EXAMINATION OF CLINICAL SPECIMENS

Direct EM examination of clinical specimens, using the negative staining tech-nique, provides the simplest and most rapid method for virus detection. Thetechnique consists of mixing a drop of specimen with a drop of heavy metal stainingsolution. The mixture is quickly air-dried on the support film covering the EMspecimen "grid," and is ready for examination within 1-2 minutes. The metalprovides an electron-dense background ("negative stain"), in sharp contrast to themore electron-transparent virus particles. The success of this method for virusdetection depends almost entirely on the concentration of virus particles in thespecimen. The quantity of material that can be examined on a specimen grid is sosmall that, in general, a positive detection requires at least 107 particles per milliliterin the original specimen [2]. Unfortunately many clinical specimens do not containthis quantity of virus; consequently, they are inappropriate for EM examination.

Undoubtedly the best type of clinical specimen for rapid virus diagnosis comesfrom vesicular eruptions associated with poxvirus or herpetic infections. The cells atthe base of the lesions produce large quantities of virus, which can be found by EM invesicle fluid or scrapings [3,4]. The large brick-shaped poxvirus can readily bedistinguished from the smaller icosahedral herpesvirus (Plate 1), and this differencein morphology, combined with the speed of the negative staining technique, has longmade EM the method of choice in the differential diagnosis of smallpox and chickenpox [5-7].Another type of clinical specimen which often contains a high concentration of

virus is the stool specimen. EM examination of stools is useful in the detection ofrotaviruses and other agents associated with acute gastroenteritis [8-12], enterovi-ruses [13], adenoviruses [8,13], coronaviruses [14], hepatitis A virus [15], etc.Nasopharyngeal secretions may contain detectable quantities of parainfluenza virus,respiratory syncytial virus, coronavirus, and mycoplasma [16,17]. Herpesviruses andmumps virus have been found by EM in cerebrospinal fluid [16,18], and CMV andBK virus in urine [19-21].

Because of the growing number of promising reports concerning the effectivenessof antiviral chemotherapy at the early stages of herpes encephalitis, the viruslaboratory is called on increasingly to perform virus studies on brain biopsies. If thesample has been correctly collected from an infected area, virus particles can usuallybe detected by EM, using the negative staining technique. A more dependableprocedure, especially with low levels of virus, is to fix and embed the tissue by one ofthe newer rapid embedding techniques [22] and systematically examine sections inthe EM. Sections are also useful in confirming a viral etiology in diseases such assubacute sclerosing panencephalitis (SSPE) and progressive multifocal leukoen-cephalopathy (PML) [2].

TISSUE CULTURE ISOLATION; EM IDENTIFICATION

Although EM examination of the clinical specimen offers the fastest method ofdetecting a virus, many specimens received by a virus laboratory may containinsufficient virus for EM detection. For this reason, EM-equipped virus laboratoriesreceiving large numbers of specimens may elect to perform direct examination only

21

TABLE IEarliest Microscopic Detection of Virus in Cell Culture following Inoculation with 100 TCID50 of Virus

Earliest detection (days)

Virus By LM* By EM**

Adenovirus 4 3Herpes simplex 3 3Coxsackie B5 2 1Parainfluenza 4 2Measles 3 3

*lnoculated cultures were examined daily by light microscopy for a cytopathic effect indicative of virusinfection.**A small sample of cells and fluid was removed daily from each inoculated culture, negative stained,and examined by EM for the presence of virus.

on selected specimens, such as biopsies, vesicle fluid or scrapings, and stools fromacute gastroenteritis, and inoculate all other specimens into cell cultures. Those viralisolates that cannot be identified by cytopathic effect (CPE) alone can be identifiedon the basis of morphology by negative staining an aliquot of the infected culture(Plate 2). As shown in Table 1, viral isolates can often be identified by EM 24-48hours before they produce a CPE that is visible by light microscopy.

VIRUS SEROTYPING BY IMMUNOELECTRON MICROSCOPY

Identification of a virus by morphology alone allows it to be classified with respectto family or group (e.g., picornavirus, adenovirus, herpesvirus, poxvirus, orthomyx-ovirus, paramyxovirus). This is usually all that is required by the clinician in order todictate appropriate treatment of the patient. From an epidemiological point of view,however, it may be important to know the precise serotype of a virus, and thisinformation cannot be obtained from virus morphology only. This problem is wellexemplified in the case of the picornaviruses, where all members, including rhinovi-ruses and enteroviruses, have an identical morphology. Enteroviruses are commonlyencountered in a diagnostic laboratory, and their serotyping is usually performed bymeans of a virus neutralization test in cell cultures. Results are obtained within l/2 to 2weeks. The same results can be obtained within one hour if one uses immunoelectronmicroscopy (IEM), by which the virus- antiserum mixtures are negatively stained andexamined on an EM specimen grid. In the electron microscope a virus-antibodycomplex appears as an aggregate of virus held together by antibody molecules (Plate3). Although still in the development stage, serotyping by IEM has been successfullyapplied not only to enteroviruses [23,24], but also to adenoviruses, papovaviruses,and myxoviruses [21,25-30].

IEM has also been used to increase the sensitivity of detection of virus in a sample.As little as 1035 TCID50 / ml of poliovirus can be detected by IEM-approximately100 times less than that needed for EM detection in the absence of antibody [23].The ability of specific antibody to form visible aggregates of homologous virus has

been utilized to detect unknown viruses that have remained elusive either by virtue oftheir indistinctive morphology, as in the case of rubella virus [31], or because theyhave been difficult or impossible to culture, as in the case of viruses such as hepatitisA and B, and wart virus [15,32-35]. A clinical specimen is mixed with the patient's

22 FRANCES W. DOANE

VIRUS MORPHOLOGY IN RAPID DIAGNOSIS 23

serum, on the theory that any virus present in the specimen may be aggregated byantibody in the serum. The mixture is then negatively stained and examined by EM.

MORPHOLOGICAL RECOGNITION OF ADVENTITIOUS AGENTS

Virology in all its forms relies heavily on cell cultures, which unfortunately maybecome contaminated with adventitious agents. One of the most notorious groups ofcontaminants comprise the simian viruses, which may occur in over 50 percent of"normal" primary monkey kidney cell cultures {36]. Established cell lines may beplagued with mycoplasma contamination, or with low-grade viral infections acquiredby cross-contamination from infected cultures in the laboratory. Such contaminantsare often extremely difficult to detect, as they may produce little or no cytopathiceffect. They can be seen by electron microscopy, however, using methods such asnegative staining and thin sectioning [37]. Virus pools passed in contaminatedcultures can themselves become contaminated. A quick check by electron microscopyprovides a means of monitoring pools for possible contaminants.

CONCLUSION

Although many virus diagnostic laboratories continue to rely heavily on the well-established isolation and identification procedures, increasing attention is being paidto the development of more rapid techniques, the majority of them involvingantibody-labeling. The fluorescent antibody technique, when adequately controlled,can be used to detect viral antigens in clinical specimens and inoculated host systems.Radioimmunoassay (RIA) and enzyme-linked immunosorbent assay (ELISA) cannow be applied with confidence for rapid quantitation of selected viral antigens andantibodies. Yet with all of these techniques the virologist must make a preliminarydecision concerning the possible nature of the causative agent in order to select theappropriate reference antisera. A major advantage of using electron microscopy forrapid virus diagnosis is that one can actually see the virus and identify it by itsmorphology. Subsequent serotyping, if required, can then be carried out moreprecisely. Serotyping by IEM is extremely rapid, and offers promise for the future. Itsvalue may increase through the use of more purified monospecific antisera.The EM techniques discussed above have all made use of the standard trans-

mission electron microscope (TEM). There are indications that the scanning electronmicroscope (SEM) may also eventually be of use to the diagnostic virologist. Atpresent, its practical application, albeit a valuable one, is limited to screening cellcultures for mycoplasma contamination [38].

REFERENCES

1. Doane FW: Identification of Viruses by Immunoelectron Microscopy. In Viral Immunodiagnosis.Edited by E Kurstak, R Morisset. New York, Academic Press, 1974, pp 237-255

2. Doane FW, Anderson N: Electron and Immunoelectron Microscopy. In Comparative Diagnosis ofViral Diseases. Edited by E Kurstak, C Kurstak. New York, Academic Press, 1977, Vol 11, part B, Ch13, pp 505-539

3. Nagler FPO, Rake G: The use of the electron microscope in diagnosis of Variola, Vaccinia andVaricella. J Bacteriol 55:45-51, 1948

4. van Rooyen CE, Scott GD: Smallpox diagnosis with special reference to EM. Can J Public Health39:467-477, 1948

5. Peters D, Nielsen G, Bayer ME: Reliability of the rapid EM diagnosis of smallpox. Dtsch MedWochenschr 87:2240, 1962

24 FRANCES W. DOANE

6. Nagington J: Electron microscopy in differential diagnosis of poxvirus infections. Br Med J2:1499-1500, 1964

7. Cruickshank JG, Bedson HS, Watson DH: Electron microscopy in the rapid diagnosis of smallpox.Lancet 2:527-530, 1966

8. Flewett TH, Bryden AS, Davies H: Diagnostic electron microscopy of faeces. 1. The viral flora of thefaeces as seen by EM. J Clin Pathol 27:603-608, 1974a

9. Flewett TH, Davies H, Bryden AS, et al: Diagnostic EM of faeces. I1. Acute gastroenteritis associatedwith reovirus-like particles. J Clin Pathol 27:608-614, 1974b

10. Middleton PJ: Clinical Observations and Diagnosis of Gastroenteritis. In Comparative Diagnosis ofViral Diseases. Edited by E Kurstak, C Kurstak. New York, Academic Press, 1977, Vol 1, part A, ChI 1, pp 423-445

11. Madeley CR, Cosgrove BP: 28 nm particles in faeces in infantile gastroenteritis. Lancet 2:451-452,1975

12. Madeley CR: Comparison of the features of astroviruses and caliciviruses seen in samples of faeces byelectron microscopy. J Infect Dis 139:519-523, 1979

13. Anderson N, Doane FW: Agar diffusion method for negative staining of microbial suspensions in saltsolutions. Appl Microbiol 24:495-496, 1972

14. Caul EO, Paver WK, Clarke SKR: Coronavirus particles in faeces from patients with gastroenteritis.Lancet 1: 1192, 1975

15. Feinstone SM, Kapikian AZ, Purcell RH: Hepatitis A: detection by immune electron microscopy of avirus-like antigen associated with acute illness. Science 182:1026-1028, 1973

16. Doane FW, Anderson N, Chatiyanonda K, et al: Rapid laboratory diagnosis of paramyxovirusinfections by electron microscopy. Lancet 2:751-753, 1967

17. Joncas JH, Berthiaume L, Williams R, et al: Diagnosis of viral respiratory infections by electronmicroscopy. Lancet 1:956-959, 1969

18. Evans AS, Melnick JL: Electron microscope studies of the vesicle and spinal fluids from a case ofHerpes Zoster. Proc Soc Exp Biol Med 71:283-286, 1949

19. Montplaisir S, Belloncik S, Leduc NP, et al: Electron microscopy in the rapid diagnosis ofcytomegalovirus: Ultrastructural observation and comparison of methods of diagnosis. J Infect Dis125:533-538, 1972

20. Lee FK, Nahmias AJ, Stagno S: Rapid diagnosis of cytomegalovirus infection in infants by electronmicroscopy. N Eng J Med 299:1266-1270, 1978

21. Gardner SD, Field AM, Coleman DV, et al: New human papovavirus (B.K.) isolated from urine afterrenal transplantation. Lancet 1: 1253-1257, 1971

22. Doane FW, Anderson N, Chao J, et al: Two-hour embedding procedure for intracellular detection ofviruses by electron microscopy. Appl Microbiol 27:407-410, 1974

23. Anderson N, Doane FW: Specific identification of enteroviruses by immuno-electron microscopyusing a serum-in-agar diffusion method. Can J Microbiol 19:585-589, 1973

24. Petrovicova A, Juck AS: Serotyping of coxsackieviruses by immune electron microscopy. Acta Virol21:165-167, 1977

25. Luton P: Rapid adenovirus typing by immunoelectron microscopy. J Clin Pathol 26:914-917, 197326. Vassall JH, Ray CG: Serotyping of adenoviruses using immune electron microscopy. Appl Microbiol

28:623-627, 197427. Edwards EA, Valters WA, Boehm LG, et al: Visualization by immune electron microscopy of viruses

associated with acute respiratory disease. J Immunol Methods 8:159-167, 197528. Penney JB, Narayan 0: Studies of the antigenic relationships of the new human papovaviruses by

electron microscopy agglutination. Infect Immun 8:299-300, 197329. Penney JB, Weiner LP, Herndon RM, et al: Virions from progressive multifocal leukoencephalop-

athy: rapid serological identification by electron microscopy. Science 178:60-62, 197230. Kelen AE, McLeod DA: Differentiation of Myxoviruses by Electronmicroscopy and Immunoelec-

tronmicroscopy. In Viral Immunodiagnosis. Edited by E Kurstak, R Morisset. New York, AcademicPress, 1974, pp 257-275

31. Best JM, Banatvala JE, Almeida JD, et al: Morphological characteristics of rubella virus. Lancet2:237-239, 1967

32. Bayer ME, Blumberg BS, Werner B: Particles associated with Australia antigen in the sera of patientswith leukemia, Down's syndrome and hepatitis. Nature (London) 218:1057-1059, 1968

33. Almeida JD, Goffe AP: Antibody to wart virus in human sera demonstrated by electron microscopyand precipitin tests. Lancet 2:1205-1207, 1965

VIRUS MORPHOLOGY IN RAPID DIAGNOSIS 25

34. Kapikian AZ, Feinstone SM, Purcell RH, et al: Detection and identification by immune electronmicroscopy of fastidious agents associated with respiratory illness, acute nonbacterial gastroenteritis,and hepatitis A. Persp in Virol 9:9-47, 1975

35. Kapikian AZ, Dienstag JL, Purcell RH: Immune electron microscopy as a method for the detection,identification, and characterization of agents not cultivable in an in vitro system. Manual of ClinicalImmunology. Edited by NR Rose, H Friedman. Amer Soc Microbiol, 1976, pp 467-480

36. Hsiung GD: Latent virus infections in primate tissues with special reference to simian viruses.Bacteriol Rev 32:185-205, 1968

37. Anderson N, Doane FW: Microscopic detection of adventitious viruses in cell cultures. Can JMicrobiol 18:299-304, 1972

38. Doane FW, Anderson N: A survey of viral cytopathology by scanning electron microscopy. ProcMicrosc Soc Can 2:58-59, 1975