Talking about Electron Microscopy of Foods

with Miloš Kaláb

From his early experiments to develop a wiener imitation from milk powder to his recent work on microorganisms, blood cells, and vegetables 23 years after his retirement

Updated: December 4, 2017.
New information

December 15, 2017.
Contributions from the following dates may be viewed:
Dec. 15, 2017
Sep. 21, 2015
Apr. 12, 2015
Mar. 12, 2015
Mar. 11, 2014
Nov. 16, 2012
Aug. 30, 2012
Oct. 2, 2011
Nov. 25, 2009
Jan. 30, 2009
May 10, 2007
Feb. 28, 2007
July 20, 2006
March 27, 2006
March 21, 2006
October 12, 2005
October 6, 2005
February 6, 2005
January 31, 2005
January 25, 2005
January 24, 2005
January 23, 2005
January 21, 2005
January 19, 2005

Learn more about the subjects mentioned using the links below.
Any comments? Please send them to
Milos Kaláb, and mark the subject line "Microscopy comment". Thank you.

Antonie van Leeuwenhoek (1632 - 1723)
Robert Koch - one of the two fathers of modern bacteriology
The life and times of Louis Pasteur.
Paul de Kruif - a man of science
Escherichia coli are common in the intestines of humans and animals and if found in foods, they indicate  fecal contamination. Very dangerous O157: H7 and similar mutants can cause severe infections, even death.
Saccharomyces cerevisiae ferment sugar and produce carbon dioxide and alcohol. They are used to make leavened bread, beer, wine and other spirits.
Light (a stream of photons) is used in optical microscopes to magnify the subject. A stream of electrons is used for this purpose in electron microscopes.
The first transmission electron microscope was developed between 1937 and 1939 by two postgraduate students working in the Physics Department of the University of Toronto with their physics professor.
Electron microscopy explained
TEM and SEM compared. Links to other kinds of microscopy.
There are two basic modes of electron microscopy (EM): scanning and transmission. Specimens destined for EM must not release gases or vapours during examination
Environmental scanning electron microscopy (ESEM) makes it possible to examine hydrated (water-containing) specimens if the water is confined to cells. Excellent images are obtained with fresh insects and plants or their parts. ESEM is, however, not suited for specimens which contain free water, such as cheeses, meat products, juicy fruits, etc.
Scanning tunneling microscopy, Atomic force microscopy, and other techniques to study the microstructure of surfaces.
Food Science.
Microstructural Principles of Food Processing and Engineering - a book by Jose M. Aguilera and David W. Stanley.
Structure of Dairy Products - a book by Adnan Tamime.
Viewers interested in microscopic images of various subjects may wish to visit some of the following sites which specialize in science:

Best Microscope Photos of the Year 2016, - 10 Amazing Microscopic Images from the Natural World, White Blood Cells Vs. Bacteria, etc.

Microscopy Websites compiled by the National Health Museum, Resource Center.

Dennis Kunkel Microscopy, Inc. The science image library holds over 1,500 light microscopy images and electron microscopy images (colorized and black & white).

Molecular Expressions TM offer one of the Web's largest collections of optical microscope images (micrographs) in colour.

MicroAngela Electron Microscope Image Gallery. Explore familiar and unexpected views of the microscopic world with colourized images from electron microscopes at the University of Hawaii. This site has been archived by the Wayback Machine, which has now over 308 billion web pages in its archive.

Electron Microscopy Black and white transmission electron microscopy images of normal cells and organs and neoplasms. Please click on the items in the "Structure" column.

Microscopy UK Resources Links to various resources

David Scharf's Images The author is known in the scientific and photographic commu- nities for his work which has been shown in numerous magazines and museums.

There are also other commercial sites called photo banks which have large quantities of micrographs in their searchable archives, e.g.:

Science Source has 4 collections, each with a large collection of micrographs:
and Videos.

Visuals Unlimited is a leading source with a wide variety of images including micrographs.

Science Photo Library presents special photo techniques, images in action, press releases, feature stories etc.

Shutterstock advertizes 176 millions of images, video clips, and music tracks.

iStock by Getty images etc.

Complete book of the microscope. Step-by-step microscope activities for children include preparing slides and observing everyday objects, insects and even your own cheek cells.

This website has been re-organized to show the most recent text at the top followed by older contributions.

Since this site had been last updated on September 21, 2015, the images of vegetables have been published as illustrations of a paper entitled Colourful vegetables - colourful micrographs. Then the author has joined ResearchGate, which is a network of researchers whose achievements may be accessed by registered members. The registration is free for anyone. At present, he has submitted a new manuscript entitled "A bowl of rice and SEM" on the microscopy of rice again for publication in "infocus". Some of the illustrations before colouring have already been in this column.

In addition to studying rice and other agricultural products, the author also collaboratored with a group of scientists led by Dr. Sandra Ramirez-Arcos at the Canadian Blood Services, particularly Dr. Mariam Taha and Dr. Maria Loza. Some of his notes in this column may, thus, be related to various subjects.

In the past year, the autor also gathered information on his scientific and technical papers to be listed in his ResearchGate profile and spent time featuring the journal Food Structure (1982-1993) in Wikipedia. When his younger colleague, Dr. D.J. McMahon, digitized all papers which have been published in the journal in order to make them freely available to all people interested in food structure, the author has facilitated the access. Readers may search for papers of interest by authors or by subjects in the Tables of Contents (or here) and then download them from the Utah State University Libraries.

Visitors to this site may also wish to search the home page for subjects of interest.

September 21, 2015.
Does the presence of a plasticizer in a sample cause problems to SEM?
Biofilms composed of human blood platelets (thrombocytes) and bacteria such as Staphylococcus epidermidis may be studied as they form on the walls of the transfusion bags. Sets of six 1 cm2 samples (called "coupons") were fixed in a buffered glutaraldehyde solution and dehydrated in a graded ethanol series. For transportation, the samples were stored in 70% ethanol for up to 3 days, then they were dehydrated in 100% ethanol and critical-point dried. A phenomenon not observed with other samples occurred when the coupons were at 40°C in the pressurized cell and CO2 gas was slowly released to complete critical-point drying. As the pressure decreased to approx. 1200 psi, the light passing through the pressurized cell became gradually obscured as if a thick fog developed inside the cell. Releasing more gas and reducing the pressure below 1000 psi cleared the cell completely.

There was a question as to whether the absolute ethanol was completely replaced with liquid carbon dioxide and whether no other volatile substance contaminated it. Decanting 70% ethanol from all samples and allowing it to evaporate has left a liquid residue which did not evaporate at 22°C in 3 days. A similar residue was obtained by evaporating absolute ethanol used in the last step when dehydrating the samples. It was assumed that the contaminant may have been a plasticizer which is used in the manufacture of some transfusion bags.

Apparently, the plasticizer vapour condenses in the form of minute droplets and forms a fog as the pressure in the pressurized cell is reduced. A small part of the fog may then settle on all surfaces in the cell including the specimens that had been critical-point dried and may affect the SEM images of the cells (platelets, bacteria). So far, no ill effects have been observed, but the phenomenon deserves a further study.

March 12, 2015.
Following confocal laser scanning microscopy of vegetables, difficulties were experienced when vivid micrographs of fluorescing constituents were prepared for printing on paper. This was done by converting them in Adobe Photoshop from the RGB mode to the CMYK mode.
"Gamut Warning" in Adobe Photoshop shows in grey which colours will not be available after the conversion from RGB to the CMYK mode.
The micrographs shown in this article are all in the RGB (red-green-blue) mode. Most commercial printing processes work with images in another, CMYK, mode (cyan-magenta-yellow-black mode. The conversion (e.g., in Adobe Photoshop) introduces considerable changes which may be seen here. Unlike graphic artists, who may start working on their images in the CMYK mode, photographic cameras produce images in the RGB mode. The loss of bright colours by the conversion is particularly clearly noticeable with the images of fluorescence, e.g., autofluorescence in a radish root shown by K. Hubbard above (left image in RGB mode, right image in CMYK mode). Adjusting the colours in the CMYK mode may bring the images somewhat closer to the originals but the fact bright colours are not available in the CMYK mode does not allow a perfect match. Blue and green colors are affected the most. (How to change colours in an image). A preview of the colours affect in available in Adobe Photoshop in View - Gamut Warning.

March 11, 2014.
Ms. Denise Chabot, Chief Technician at EMU, and Technician Keith Hubbard have expertise in all kinds of electron microscopy and also in confocal laser scanning microscopy (CSLM). Their colourful micrographs of vegetables are shown below.
Autofluorescence CLSM image of a section from a carrot centre by D. Chabot.
In the past several years, confocal laser scanning microscopy (CLSM) has figured prominently in the activities of the EMU staff. M. Kaláb gladly accepted the assistance and advice from Ms. Denise Chabot, the new Chief Technician, particularly in his new interest in the microstructure of vegetables such as carrots and other vegetables.

Autofluorescence CLSM image of the skin of a raw tomato by D. Chabot.
Vegetables are an important part of human nutrition. There is a growing evidence that a diet rich in fruits and vegetables is associated with a lower risk of some malignancies. Almost all parts of plants are considered to be "vegetables", from flower buds (e.g., broccoli, cauliflower) to leaves (e.g., cabbage, kale, lettuce, spinach) to stems (e.g., ginger, kohlrabi), stem leaves (e.g., celery, rhubarb) to roots (e.g., beets, carrots, radishes) to tubers (e.g., Jerusalem artichokes, potatoes, yams) to bulbs (garlic, onions) to seeds (e.g., beans, peas, sweet corn). Interestingly, some of the "typical" vegetables mentioned above are actually fruits in the botanical sense (cucumbers, eggplants, pumpkins, squash, tomatoes etc.).

Colour-enhanced SEM image of a secton of a celery stalk by D. Chabot.
Vegetables provide macronutrients (starch in tubers such as potatoes and in seeds such as beans, lentils), or inulin in Jerusalem artichokes, garlic, onions, dandelions etc.), and also micronutrients such as minerals and vitamins, antioxidants, flavours and other substances. Dietary fibre is an important group of substances present in the vegetables which are important for human wellbeing.

Same image in CMYK mode (see the next subject on RGB vs. CMYK).
Autofluorescence CLSM image of a radish root by K. Hubbard.
The interpretation of CLSM micrographs is not easy and is, in many instances, impossible even for experts knowled­geable of plant anatomy. Unlike SEM, which visualizes surfaces (e.g., a section of a celery stalk, above), CLSM registers substances in the specimens (e.g., a carrot section or tomato fruit skin) which fluoresce in specific areas of the ultra­violet (UV) light spectrum and display a variety of colours. A micrograph of a radish root obtained by K. Hubbard is such an example.

Corn seed section stained for CLSM using Acridine and Calcofluor to induce fluorescence of the seed constituents by K. Hubbard.
In contrast, a corn grain section stained with Acridine Orange (for DNA) and Calcofluor White for cell walls displays a well organized cellular structure.

Brightfield microscopy of hard-boiled egg yolk by K. Hubbard. Bar: 200 µm.
A food not examined by optical or elec­tron microscopy earlier on this page is hard-boiled egg yolk. Granularity of the boiled egg yolk may be noticed by a naked eye. All egg lipids are actually concentrated in the egg yolk. The grains may be separated from each other and dispersed in vegetable oil. Light micro­scopy at right shows them as polyhe­drons resulting from their tight packing in the yolk.

November 16, 2012.

Mr. Ann-Fook Yang retired in 2012

Mr. Ann-Fook Yang (in the front) was the Chief Technician at the Electron Microscopy Unit (EMU), Agriculture and Agri-Food Canada in Ottawa until he retired in November 2012 after 34 years of work in that position. In the snap­shot at left, the new Chief Technician,, Ms. Denise Chabot, points to an image on an FEI Quanta 600 ESEM microscope while Technician Mr. Keith Hubbard stands behind. Miloš Kaláb looks from the back.

Mr. Yang will be remembered for many good deeds at EMU, for his assistance to many scientists and technicians and co-authorship of scientific papers and his own work, and also for two important acquisitions – a Hitachi 7000 transmission electron microscope and the FEI Quanta 600 ESEM.

August 30, 2012.
SEM of toxigenic Escherichia coli bacteria was used to evaluate their dimensions.

SEM images of six  strains of toxigenic (hemorrhagic) E. coli bacteria (O26, O45, O103, O111, O121, and O145) studied by Ms. Kyriaki Chatzikyriakidou at the University of Wiscon­sin were obtained in sufficient quantities to make statistical evaluation of their dimensions possible. Characteristic images (at left) show differences expressed also in the form of a diagram (below). The lengths of the bacteria are expressed in micro­meters with the scale at left.

Ms. Chatzikyriakidou defended her thesis entitled "Behavior of Shiga-toxin-producing Escherichia coli (STEC) in beef and beef model systems under cold and acid stress" in December 2014 at the University of Wiscon­sin - Madison: "In the past decade, non-O157 Shiga toxin-producing Escherichia coli (STEC) strains have been implicated in an increasing number of foodborne outbreaks in United States of America, especially those related to beef, adding to the numbers of outbreaks caused by the well-known E. coli O157:H7. This research portfolio firstly compared the survival and growth of non-O157 STEC under cold- and acid- stress, and secondly explored the attachment of a variety of STEC strains to bovine collagen type I and myofibrils".

Now, interested in student psychology, she is doing biology education research at the same university, where she teaches a couple of classes for entry level research. In addition, she collects microorganisms in caves and photographs them using an optical microscope.
(Updated April 12, 2015.)

October 2, 2011.
After many years of examining milk products by electron microscopy, the major findings have been summarized in a paper entitled The Beauty of Milk at High Magnification (infocus, The Proceedings of the Royal Microscopical Society, issue 18 June 2010, pp. 4-37).

A new paper entitled Microscopy and Hygiene was published in the September 2011 issue, pp. 4-32. It is now available also on the Internet.

      Micrographs in both papers and also in the review entitled Conventional Scanning Electron Microscopy of Bacteria by M. Kaláb, A. F. Yang, and Denise Chabot (June 2008, pp. 42-61) have all been rendered in false colours. The cover page of the infocus June 2010 issue (at left) shows processed cheese (Fig. 17 in the paper) where fat globules are seen in the process of emulsi­fication. A freshly developed calcium phosphate crystal is shown in blue colour.
      The Royal Microscopical Society in the United Kingdom welcomes submission of articles of general interest to micros­copists. Full articles or reviews ideally between 3000 and 4000 words contain images/figures, where relevant. There is no limit of the number of colour images. All micrographs must include scale bars or field widths and are expected to be submitted in CMYK colour mode. To view the full guidelines for authors, please visit
      As the author, I thank Scientific Editor, Dr. Adrian Burden, and Editor, Ms. Laura Kingsbury, and her team for the beautiful form which they have given to my three contributions. They may be contacted at

November 25, 2009.

The food we eat eventually leaves our body after having been digested and depleted of nutrients. How has it changed during that process? Profoundly.

        Digestion breaks down all food constituents. Although our body extracts most nutrients, there is still enough of them to support microorganism in the lower digestive tract, i.e., the large intestine. In fact, some bacteria are even beneficial for our health. A wide variety of bacteria represent about one third of the volume of the intestinal contents. This subject was already mentioned in May, 2007. To view the bacteria under the SEM, the excrements need to be prepared in the same way as other samples but dry fracturing would be preferred to freeze fracturing.
        Animal feces and bird droppings contain microbial pathogens which appear in our environment due to the lack of hygienic measures. The bacterium Escherichia coli is present in the intestinal contents of most animals and is thus called the "intestinal rod". When it is found in other environments, it is an indicator that fecal matter is present.
        The swine flu epidemic in 2009 has renewed the interest in protecting our health. SEM images of unhygienic conditions may, therefore, be educational.
An imprint of a dirty human hand on a nutrient agar plate. Bacterial colonies developed where bacteria were tranferred from the hand onto the agar gel.
Bacteria are present in an old toothbrush along with decaying matter. Bar: 2 µm.
        Do we handle food with clean hands? An unwashed hand imprinted on an agar plate with a nutrient medium created an image of the hand within several days as the bacteria transferred onto the plate developed into colonies. The image thus indirectly shows the contamination of the hand with bacteria.
        The foods which we bring home from a foodstore also contain bacteria. Leaving a cut of meat to lie at room temper­ature instead of placing it into a refrigerator ensures that bacteria will grow on it and make it slimy and unfit for human consumption.
        Food enters our mouth which is already full of its own bacteria - on the tongue and on the teeth. Older people clean their tongue and everyone brushes one's teeth. In spite of washing the toothbrush too, bacteria take hold around the bristles and propagate, forcing us eventually to replace the toothbrush.
        The role of bacteria in the digestive system has been the subject of scientific studies. Nowadays the tendency is to colonize the system with probiotic, including lactic acid bacteria and bifidobacteria.

SEM of an aerosol droplet dried on the wall of a toilet bowl. Bar: 0.2 mm.
Bacteria - mostly bacilli (rods) - in Canada goose droppings. Bar: 2 µm.
        In a civilized society, human feces are flushed with water into a sewage system designed to decontaminate the water and to compost the solid matter. A small part of the toilet bowl contents becomes airborn in the form of aerosol droplets. The droplets settle on the bowl walls and may travel out of it if the lid is not closed down. Microscope cover slips attached to the bowl walls for several days reveal the dry matter of the droplets. Some bacteria were found at a high magnification but many were obscured by the fecal solids.

Canada goose droppings in grass.
        There are various sources of fecal matter in our environment which cause the closing of public beaches after a heavy rain which flushes animal and bird excrements into the river or lake. In and around Ottawa and certainly elsewhere in North America, Canada geese foul lawns, playgrounds, sidewalks, and river water with their droppings of only partially digested grass. Because of high local concentrations of the birds, bacteria in their droppings are a potential health hazard. The situation is aggravated by visitors to municipal parks who feed wildlife in spite of many signs advising them to resist that urge.

January 30, 2009.

    Preliminary coating of bacterial filters with gold may also be used before the bacteria are captured on the filter to prevent potential charging arefacts if the success of SEM is in jeopardy.

Charging artefacts in an SEM micrograph     Proper preparation of specimens is essential to obtaining excellent SEM micrographs. The need to work with very small specimens (~1.5 mm in diameter) was emphasized in various connections, particularly if such specimens need to be fixed and dehydrated. However, even small dry particles may be quite difficult to prepare properly for SEM. Charging artefacts at left (white arrows; bar=100 µm) are probably the best known problem. They originate when the conductive gold coating on the specimen has gaps. Electrons hitting the specimen cannot be discharged and their charge consequently deflects the incident electron beam thus creating the artefact. This has already been mentioned above in this webpage.
    A double-sided sticky tape has been used to attach small particles to the SEM stub. Additional painting of the particle side walls with a conductive silver cement improves the contact between the particle and the support. However, powder and dust particles are often very small (<10 µm) and they sink into the dense sticky material of the sticky tape.

Minute particles on sticky stuff Minute particles on gold-coated cover slip

    One possibility to alleviate this problem is to dissolve the sticky material in an organic solvent such as acetone or toluene and to spread a droplet over a glass cover slip, 13 mm in diameter. Evaporation of the solvent leaves a considerably thinner sticky layer which may provide a good contact without submerging the particles too deep (left image, bar: 200 µm).
    Another possibility is to avoid the use of any binding substance. The glass cover slip mounted on an SEM stub is sputter-coated with gold while a small amount of the powder (such as very small pollen grains, plant protein bodies, starch granule fragments) spread in a single layer on another cover slip is only half-coated with gold (10 nm), the coating is stopped, the particles are stirred and spread again, and gold-coating is completed. Then the particles are transferred onto the gold-coated cover slip and a droplet of absolute ethanol is used to spread the particles over the entire cover slip. Ethanol is then allowed to evaporate and the particles are coated with a thin layer of gold (~10 nm). This procedure ensures that there will be an electrically conductive contact between the particles and the support so that charging artifacts will not develop (right image, bar: 200 µm).

May 10, 2007.

    The possibility that leaf vegetables, which are consumed raw, may be contaminated with the feces of wild animals poses a health hazard. It is aggravated by the fact that fecal bacteria enter the leaf stomata and may not be washed out with water.

Escherichia coli bacteria were found as regular harmless inhabitants of the human intestines in 1885 but one hundred years later, some of these bacteria were found to produce Shiga-like toxins called verotoxins. Such bacteria are called toxigenic and are marked as Escherichia coli O157:H7. Lately they caused food poisoning in North America when they appeared on leafy vegetables which are consumed raw, such as lettuce and spinach.

SEM was used to visualize the presence of E. coli on the leaves. For illustration purposes, live nonpathogenic E. coli at very low concentrations were reacted with lettuce and spinach leaves and then fixed and photographed. It is interesting to note that particularly on lettuce leaves the bacteria congregate at leaf stomata and even enter them, thus hiding inside the leaves. It is evident that it would be impossible to wash the bacteria out from the stomata. Compared to lettuce leaves, spinach leaves (image at right taken at a lower magnification) are considerably more convoluted and the bacteria may hide in the convolutions. It would be equally impossible to wash them out from their hiding places. Their nature, i.e., their adaptation to life in the intestines makes it very easy for them to propagate in the intestines if they are ingested with the leaves in the form of raw salads.

Bacteria may not be identified from their SEM images, like in the micrograph of human feces shown at left but this micrograph shows how plentiful the bacteria are in the excrements and what proportion of the entire mass they represent. It is no wonder that a small piece of an excrement may contaminate a large volume of water. The micrograph also indicates that it is very important to thoroughly wash one's hands after the use of a toilet.

February 28, 2007.

    Anaglyphs are pairs of superimposed stereograms, one in red and the other in green, cyan, or blue, which give the impression of a three dimensional object when viewed through a pair of glasses of corresponding colours.
    Anaglyphs are easy to make using a scanning electron microscope even at very high magnifications.

    The image at left is an anaglyph - a stereogram. When viewed through a pair of red-and-green glasses, this image will reveal the three dimensional structure of the yeast cells and bacteria on the surface of a kefir grain.
   At Christmas in 2006, my 7-years old grandson Adam showed me a book full of photographs of spiders, where many of them were in the form of anaglyphs. Using a pair of red-and-green glasses it was possible to see the spiders as 3-dimensional creatures. Aerial photographs of mountains in museums have also been often shown in the form of anaglyphs.
    In the past, when images in colour were extremely rare in food science journals, the 3-dimensional effect was obtained by printing pairs of micrographs which had been taken at a difference of 6 to 12 angular degrees. The easiest way to view them was through a simple optical apparatus but many people have learned to obtain the 3-dimensional effect simply by focussing their eyes behind the journal's page.
    There is free software on the Internet and that has made the task easy. Examples of SEM anaglyphs and guidance how to make them may be accessed here. Anaglyphs enhance the ability of the scanning electron microscope and make it easier to evaluate vertical distribution of structural elements, for example bacteria, spores, blood cells etc. Many interesting anaglyphs may be found on the Internet, e.g., here, here, here, here, as an image of the bearded stonewort (Chara canescens), or in Fig. 13 in a paper on the SEM of bacteria.

July 20, 2006.

Negative staining may be claimed to be one of the simplest TEM techniques but depending on the subject studied, it may appear to be quite demanding. There are several critical points such as:
    1. The carbon-coated grids must be made hydrophilic.
    2. Casein micelles must be preserved by fixation and the presence of a dilute calcium chloride solution.
    3. The micelles must be separated from whey proteins by ultracentrifugation and then redispersed.
    4. Proper ratios of the micelles and the negative stain must be used and their excess removed.
    5. The remaining thin layer of the liquid must be freeze-dried to retain the globular shapes of the micelle.

  Recently I have received several requests for advice on negative staining of casein micelles for subsequent transmission electron microscopy (TEM). The writers complained that they have been obtaining fuzzy images. Regrettably I have been unable to find exemplary micrographs of negatively stained casein micelles on the Internet. However, I reprinted images of negatively stained casein micelles (obtained from E. Uusi-Rauva at the State Control Office for Dairy Products in Helsinki, Finland) in 1977 in an article entitled "Electron Microscopy in Dairy Research" (MSC-SMC Bulletin, November 1977, pp. 4-10 - i.e. the Microscopical Society of Canada Bulletin). Then, 4 years later, I published my own micrographs (see below).
    The text from my own publication (Electron Microscopy of Milk Products: A Review of Techniques by M. Kalab, Scanning Electron Microscopy 1981/III, 453-472 - SEM Inc., AMF O'Hare (Chicago), IL 60666, USA) reads as follows:

2. Transmission Electron Microscopy

    TEM comprises all techniques in which the specimen is placed in the electron beam and the enlarged shadow is examined. There are various methods of preparing the specimen for this kind of study.

2.1. Negative Staining
    Negative staining is probably the easiest TEM technique. The specimen is in the form of a suspension of submicroscopical particles semitransparent to the electron beam. Addition of phosphotungstic acid (PTA), sodium phosphotungstate, or ammonium molybdate solutions (Ref. 64) to the suspension makes the medium but not the particles, electron-dense. After a thin layer of the suspension is dried, the electron beam passes only through the semitransparent particles under study and is absorbed by the surrounding stain. The particles appear light against a dark background in the micrographs (Fig 15). 
Irradiation of Formvar- and carbon-coated grids with ultraviolet light or exposure to glow discharge (Ref. 65) make the grids wettable for aqueous suspensions. Most suspensions are diluted prior to mixing with the PTA solution to prevent overlapping of the particles. A proper pH of the mixture is also an important factor. A preliminary fixation with formaldehyde (Ref. 65, 67) or glutaraldehyde (Ref. 29) is advisable particularly with casein micelles.
    Interfering whey proteins are separated from casein micelles by ultracentrifugation and washing of the sedimented casein micelles prior to their negative staining (Ref. 68).
    A suspension under study is usually mixed with an equal volume of a 2% PTA solution and a small droplet of the mixture is placed on prepared grids. Excessive liquid is removed after several minutes by touching the droplet with a piece of a filter paper and a thin layer of the mixture left on the grid is allowed to dry for the electron microscopic examination.
    This technique is limited to dilute suspensions and, hence, has been used mostly to study the ultrastructure of casein micelles (Ref. 66, 67, 68), lipoprotein membrane fragments (Ref. 69), and bacteriophages (Ref. 70) in lactic cultures.
    As PTA penetrates porous particles such as the casein micelles, it shows their corpuscular ultrastructurfe in great detail.
    Calapaj (Ref. 66) used negative staining to study the ultrastructure of bovine and human casein micelles in fresh milk and in milk acidified with lactic acid.
    Uusi-Rauva et al. (Ref. 67) examined the effects of various pasteurization and storage temperatures on the disintegration of casein micelles into their subunits. Recently, Creamer and Matheson (Ref. 71) showed by negative staining that casein micelles increased in size with heat treatment, the effect being greater at higher pH and higher temperature.
    Surfaces of the casein micelles were found to be more diffuse after heating to 403 K (130°C) than to 373 K (100°C). Snoeren (Ref. 72) and Snoeren et al. (Ref. 73) used negative staining to study interactions of milk proteins with kappa-carrageenan. Keenan et al. (Ref. 74) found distinct morphological differences between plasma membrane and milk fat globule membrane.
    In biology, M. Nermut (Freeze-Etching Techniques and Applications E. L. Benedetti and P. Favard (eds.), 274 pp., Société Française de Microscopie Électronique, Paris, France, 1973) showed that air-drying resulted in the collapse of influenza virus particles. Freeze-drying of the negatively stained preparation (rather than air-drying) preserved the globular shapes of the particles. It is almost certain that casein micelles also become flattened and disfigured during air-drying, so freeze-drying would be highly recommended.
    Disintegration of the casein micelles during negative staining as well as during preparation for shadowing with platinum and carbon may be prevented by adding calcium chloride to the diluted milk as suggested by Rose and Colvin (Ref. 75, 76).

References related to negative staining:

29. Carrol RJ, Thompson MP, Nutting GC: Glutaraldehyde fixation of casein micelles for electron microscopy. J. Dairy Sci. 51, 1903-1908, 1968.
66. Calapaj GG: An electron microscope study of the ultrastructure of bovine and human casein micelles in fresh and acidified milk. J. Dairy Res. 35, 1-6, 1968.
67. Uusi-Rauva E, Rautavaara J-A, Antila M: Über die Einwirkung von verschiedenen Temperatur Behandlungen auf die Caseinmicellen. Eine Elektronen mikroskopische Untersuchung unter Verwendung von Negativfärbung. Meijeritieteellinen Aikakauskirja (Helsinki) 31, 15-25, 1972.
68. ShimminPD, Hill RD: Further studies on the internal structure of the casein micelles of milk. Austral. J. Dairy Technol. 20, 119-122, 1965.
69. Stewart PS, Puppione DL, Patton S: The presence of microvilli and other membrane fragments in the non-fat phase of bovine milk. Z. Zellforsch. 123, 161-167, 1972.
70. Knoop A-M: Milchforschung mit dem Elektronenmikroskop. Z. Lebensm. Unters. Forsch. 168, 305-313, 1979.
71. Creamer LK, Matheson AR: Effect of heat treatment on the proteins in pasteurized skim milk. New Zealand J. Dairy Sci. Technol. 15, 37-49, 1980.
72. Snoeren THM: Kappa-Carrageenan. A Study on Its Physico-chemical Properties, Sol-Gel Transition and Interaction with Milk Proteins. NIZO Verslagen, Ede, the Netherlands, 64-91, 1976.
73. Snoeren THM, Both P, Schmidt DG: An electron microscopic study of carrageenan and its interaction with kappa-casein. Neth. Milk Dairy J. 30, 132-141, 1976.
74. Keenan TW, Morré DJ, Olson DE, Yunghans WN, Patton S: Biochemical and morphological comparison of plasma membrane and milk fat globule membrane from bovine mammary gland. J. Cell. Biol. 44, 80-93, 1970.
75. Rose D, Colvin JR: Internal structure of micelles from bovine milk. J. Dairy Sci. 49, 352-355, 1966.
76. Rose D, Colvin JR: Appearance and size of micelles from bovine milk. J. Dairy Sci. 49, 1091-1097, 1966.

Additional information on negative staining:

March 27, 2006.

Fungus Monascus purpureus growing on rice turns the grains red. The product is called "red yeast rice". The grains are covered with minute crystals of statins, substances known to reduce blood cholesterol. Consumers of red yeast rice should be aware of its side effects, particularly, if they make it at home. It is relatively easy to prepare, if a Monascus purpureus culture or fresh red yeast rice is available as the inoculum.

     While examining long-grain rice starch by scanning electron microscopy for another scientist, I obtained specimens of many other kinds of rice from my Chinese colleague, Mr. Ann-Fook Yang. Thus I also examined black glutinous rice, short-grain rice, glutinous rice, scented rice, red jasmin rice, and red yeast rice.

     Rice starch differs by its shape from starch granules isolated from other sources such as wheat, corn, potatoes, beans etc. Whereas the latter starches produce oval shapes as cross sections of the granules, rice starch granules produce polygons (image at far left; bar: 5 µm). Rice starch is very tightly packed in the rice grain cells. The starch granules have sharper edges in long-grain rice than in glutinous rice. A fractured grain may reveal minute globular protein bodies among the starch granules or their imprints may be seen on the starch granules as minute dimples.

     Red yeast rice has a grain structure similar to that of the other rice varieties but the starch-holding cells are loosened from each other and the starch granules appear to be degraded or porous (image at left; bar: 10 µm). Red yeast rice is produced in Asia by culturing a red fungus (Monascus purpureus) on the rice grains and then drying the red product.

     The fungus propagates on the rice grains and grows its hyphae deep into the grain where they obtain nutrients. In this process, the grains cells separate from each other and the starch granules degrade. The fungus produces Monacolin K, also known as mevinolin, the same active agent (statin) as in the cholesterol-reducing drug Lovostatin. The substance may be seen in the form of plentiful minute crystals (bar: 10 µm) on the surface of the rice grains. Red yeast rice has been valued in China for many centuries as a natural cholesterol reducing herbal medicine. It has been now produced and marketed in the West, where it may have been accepted by patients who have switched to a natural substance. There are, however, some important words of caution, particularly an advice to consumers to take even red yeast rice under medical supervision because of possible side effects and to be even more careful if pharmaceutical cholesterol-lowering medicines are also taken at the same time.

March 21, 2006.

Bacteria are everywhere. Would foods be safer, if the animals which we eat (chickens, turkeys, pigs, lambs, steers) would contain only "friendly" (probiotic) bacteria which would suppress microbial pathogens? There are some experiments in progress in this direction.

     The presence of bacteria in chicken intestines may seem to be only distantly related to foods. Every living creature has bacteria in its guts, so why to be concerned?. Experts who care for the safety of foods for human consumption would prefer if the bacteria in chicken intestines be "friendly to humans" rather than pathogenic. There are indeed many "friendly" bacteria and they are called "probiotic", such as lactic acid bacteria (lactobacilli, lactococci) which utilize milk sugar lactose and oxidize it into lactic acid. This acid then coagulates milk and halps to produce sour milk, yogurt, kefir, cultured buttermilk and other dairy product. In cheeses, lactic acid bacteria also participate in their ripening. Leuconostocs are also part of lactic acid bacteria. They may be found in sauerkraut and various pickled vegetables. Bifidobacteria are another group of probiotic bacteria. Food safety experts are interested to know whether probiotic bacteria could be used to colonize the intestines and thus replace harmful pathogenic bacteria such as salmonella, staphylococci etc. Cultivation and microbiological identification would be used in such studies. Yet, it appeared interesting for an electron microscopist to find the bacteria in their niche. My objective was thus to find bacteria in the mucus which adheres to the intestinal mucosa and even directly on the intestinal villi. This means that I was not interested in the intestinal contents.

     Traditional preparation methods for SEM remove the intestinal mucus and show a clean surface of the mucosa. Thus fixation of the mucus is essential. The mucus consists of polysaccharides, so fixatives different from glutaraldehyde must be used. In their studies of intestinal mucus in young pigs, Paula Allan-Wojtas et al. (Microscopy Research and Technique 36:390-399, 1997) successfully used Ruthenium red (RR) or Alcian blue (AB) added to a glutaraldehyde fixative to retain the mucus. The authors illustrated their paper with 30 micrographs, all taken at low magnifications to show the mucus but not the microorganisms.

     In our experiments, 1 cm2 intestinal samples were excised and fixed according to Allan-Wojtas et al. Alcian blue produced beige specimens with very little mucus retained. Ruthenium red produced purple specimens and both the mucus and the resident bacteria were easy to find. The fixed samples were cut into 1 mm x 1 mm x 10 mm strips, dehydrated in ethanol, and freeze-fractured. Examination of the fractures showed both the mucus and plentiful bacteria, particularly in the caecum (see the illustration) and the colon. Various shapes of the bacteria indicated than no single species prevailed, although we are aware that bacteria cannot be identified from their shapes alone. Microbiological procedures have been used for that purpose. SEM showed bacilli (rods), cocci, clubs, spirochetes etc. during an extensive search as the micrograph above at left shows. Freeze-fracturing made it also possible to photograph the microvilli of the intestines.

October 12, 2005.

It is easy to prepare solid specimens for scanning electron microscopy. What about semisolid specimens such as coagulated milk (drinkable yogurt, cultured buttermilk), egg yolk, etc.? Encapsulation in agar gel tubes makes it possible to handle such specimens as if they are solid.

     In the summer, I received a request for SEM micrographs of cultured buttermilk and stirred yogurt from an author who wanted to illustrate his chapters on these products in his new book on cultured milk products.

     Semisolid products would disintegrate in the aqueous solution of glutaraldehyde during fixation. There is, however, a procedure whereby viscous samples may be fixed and further prepared for SEM without any problem after they are encapsulated in agar gel capsules (schematically shown at left). This procedure, initially developed in 1968 by R. K. Salyaev (Proc. 4th Europ. Reg. Conf. Electron Microsc., Rome, II, 37-38) is suitable for various suspensions and emulsions including very soft centrifugation pellets.
     The sample is first aspirated in a pasteur pipette (step 1), ~12 mm in diameter, and the end is closed with a small piece of agar gel. A small droplet of an agar sol (concentration 3-4%, temperature ~40°C) may be applied at the end of the Pasteur pipette in the case of viscous samples (step 2). Then the Pasteur pipette is briefly dipped into the agar sol and is immediately rotated to form a thin layer of gel on the glass. This step (#3) is repeated several times if necessary. Using a blade, the upper end of the gel is trimmed at the upper end of the sample (step 4). Then the gel on the pipette is gently taken into 3 fingers and the pipette is pulled away. Its contents stay in the agar gel tube (step 5). Finally, a droplet of agar sol is placed into the upper end of the agar gel tube to close the contents inside (step 6). The sample is thus encapsulated and may be handled as a solid sample. This means that the sample may be fixed, dehydrated, even freeze-fractured, and critical-point dried.
     This procedure was used with many food samples such as cultured buttermilk, drinkable yogurt, ketch­up, and prepared mustard, and also with samples such as suspensions of fixed bacteria, yeasts, and animal and bird droppings. However, there are microscopists, who consider this technique to be slow and difficult to perform. They encapsulate their samples in a different way. Grooves are made in an agar gel slab and then are filled with various samples. A layer of warm agar sol is poured over and is allowed to solidify. The solid agar gel block is then cut into smaller individual blocks (1a and 1b in the image at right). The author received such samples to examine and was not delighted. For a comparison, similar samples of coagulated milk were encapsulated using the Pasteur pipette procedure (2a and 2 b). Fixation and dehydration of these uniform cylindrical samples is more rapid and freeze fracturing produces fragments which are easier to mount and the resulting SEM micrographs are superior to the "faster encapsulation procedure".

October 6, 2005.

Information that an image in a magazine or a book is shown at a certain magnification is true only in the case that the publisher calculated it with respect to the size of the image printed. It is preferable to show a micrometer marker or to indicate the width of the area enlarged in micrometers. Projecting an image on a screen is a good example to show that the true magnification changes with the size of the image projected.

     A few days ago, a colleague asked me why I do not show bacteria on the tip of a needle, on a computer keypad, on the edge of a drinking straw and similar objects which "must be full of microorganisms". Of course, it would be instructive to show such images - other people asked similar questions. So let's look at the letters n or o on the monitor in this text. They are a little over 1 mm wide. Enlarging them 1000 times would make them 1000 mm = 1 m (about 3 feet) wide. If these letters were coated with bacteria, they would look, at that magnification, very small, like tiny dots or very short lines. To show bacteria as three-dimensional objects with flagella, I use magnifications between 8,000 and 20,000 times. The letters "n" and "o" would each be 8 to 20 metres (24 to 60 feet) wide. The bacteria would then look like the fingers of a little child. The tips of needles would look like a very wide hills - not sharp at all. The background, to which the bacteria were attached, would be irrelevant. To make viewers believe that salmonella were photographed on chicken skin, it would be necessary to assemble a montage (collage), where a micrograph would be superimposed over a photograph.

     Scientists express magnification either by stating the width of the image in micrometers (1 µm = 1/1000 mm) or by showing a bar of a certain length (1 µm, 2 µm, 5 µm, 10 µm, etc.). Lactobacillus acidophilus bacteria are shown at two magnifications. At the higher magnification (image at left), the 2 µm bar is 16 mm long on my monitor. This means, that a 1 µm long bacterium would appear 8 mm, i.e., 8000 µm long. The bacteria would, therefore appear to be magnified 8000x. If your monitor shows a different length, the true magnification can be calculated in the following way:

M = 1000A/B      where

M = the true magnification in a print, a book, on a poster
A = length of the bar in mm (as measured by a ruler)
B = value of the bar in micrometers as stated on the micrograph or in the legend.

     If the width of the image is given in micrometers, a similar formula is used to calculate the true magnification:

M = 1000D/E      where

M = the true magnification in a print, a book, on a poster
D = the width of the image as measured by a ruler
E = value of the image width as stated in the legend.

     The width D of the image above is 88 mm on my monitor. It shows an area 11 µm wide (E). The magnification M = 1000x88/11 = 8000. The same result is obtained as by calculating the magnification from the micrometer bar.

February 6, 2005.

"Charging artifacts" in SEM are caused by the accumulation of negative charge on the specimen due to a defective conductive path for the electrons. This charge deflects the incoming electrons and is manifested as light or dark lines or white areas of no structure.

     Last week, I was showing an environmental scanning electron microscope to a food scientist. He was excited to learn that such microscopes make it possible to examine hydrated specimens. "How can it be done?" he wondered, knowing well that specimens destined for electron microscopy must not release any vapours or gases while they are being examined.
Specimen stage (light arrow) in an open environmetal scanning electron microscope (Philips ESEM XL30). (The black attachment with a funnel is part of a cryo-SEM accessory).

     It is a tricky arrangement. In biological applications, the stage is cooled to around the freezing point of water (0°C) to reduce the vapour tension above the hydrated specimen and the air pressure is somewhat but not fully reduced (<9 torr) around the specimen. This space of somewhat reduced air pressure is surrounded with space where the pressure is considerably lower. The electrons thus pass through high vacuum and only the last few millimetres above the specimen they encounter higher pressure.
     The presence of water molecules in the atmosphere around the specimen makes it possible to examine specimens not coated with gold and yet, "charging artifacts" are no problem.

     "This is great news for milk products and other foods", my colleague happily exclaimed, thus repeating the reactions of other food scientists. I am sorry to cool down this excitement. Environmental scanning electron microscopes (ESEM) are excellent in structural studies of hydrated specimens in which water is enclosed in cells, such as insects or plants. Most foods, however, are not of this kind. They contain "free" water at their surfaces which covers the solid matrix composed of proteins (cheeses, meats), polysaccharides (fruits, vegetables, starch gels), fats (butter, mayonnaise) or their mixtures. Enlarging such surfaces by ESEM would not show the immersed solids. In addition, many hydrated foods also contain volatile substances other than water and may also contain free fats or oils. Such substances would soon contaminate the column of the microscope and reduce its efficiency. As I have already experienced it, subsequent examination of dry, gold-coated specimens such as microorganisms at high vacuum at much higher magnifications around 20,000x does not produce high-quality images.

Some hydrated (water-containing) specimens may be examined in environmental scanning electron microscopes (ESEM) without the need to be frozen or dried. Excellent results are obtained with specimens where the water is confined to cells such as insects or plants and their parts. It is not suitable for moist foods such as cheeses or meat products.

Fungus Stachybotrys chartarum by ESEM. Arrow points to water condensation on the spores. Bar: 10 µm.
     Food-borne microorganisms comprise bacteria and fungi (moulds and yeasts). Of these two groups, moulds are good subjects for ESEM as a micrograph of Stachybotrys chartarum shows at left, although the balance between temperature and air pressure was not ideal. Some water condensed on the spores (arrow). The opposite problem, drying, would collaps and flatten both the spores and the hyphae.
     In contrast, it is actually almost impossible to show bacteria properly because they live in aqueous media. Their surfaces may be covered with capsules which consist of polysaccharide gels, or the bacteria have a variety of minute structures such as fimbriae or flagella on their surfaces, all of which would retain free water. ESEM would show the aqueous envelope. Because of the minute size of the bacteria, it would be extremely difficult to adjust the temperature of the specimen and the air pressure above it, as these two factors determine whether water will evaporate from the sample (due to increased temperature or reduced air pressure) or will condense on it (decreased temperature and increased air pressure).
"Ropy" lactobacilli in yogurt. The filaments are artifacts which developed whe"Ropy" lactobacilli in yogurt. The filaments are artifacts which developed when the yogurt specimen was prepared for SEM.
     However, photography of bacteria prepared for high-vacuum by fixation, drying, and gold-coating is not free of artifacts. The polysaccharide gel capsules shrink into filaments on dehydration thus producing artifacts. No such filaments exist around the bacteria in their native state. In the seventies of the past century, several dairy scientists explained that some bacteria in cottage cheese use these filaments to anchor themselves to the solid casein matrix to avoid being washed away with the whey. The truth is that such bacteria (called "ropy bacteria") developed thicker capsules. Their poly- saccharides retained more water and increased the viscosity of cultured buttermilk or yogurt compared to "regular bacteria". The polysaccharides turned into filaments during the preparation of the specimens for electron microscopy. Since dehydration in alcohol is also used in the preparation of specimens for transmission electron microscopy, the filaments were shown by both techniques and, thus, were believed to be genuine.

Filaments seen "to radiate" from bacteria in SEM micrographs of yogurt and Cottage cheese are bacterial polysaccharide capsules shrunken during the preparation of the specimens for electron microscopy. They are artifacts.

January 31, 2005.

Artifacts happen - and no one is safe from them. If there is defective planning, repeating the experiments will reproduce the artifact and fool the researcher.

     One of the best procedures to obtain excellent SEM images of the internal micro- structure of foods such as yogurt, curd, cheese, and tofu consists of fixing the specimen, preferably in the form of 1x1x10 mm prisms. The prisms are then dehydrated in a graded ethanol series. When they are impregnated with absolute (100%) ethanol, they may be frozen directly in liquid nitrogen although, contrary to general belief, liquid nitrogen is a poor cryogen. If a subject is immersed into it, liquid nitrogen immediately forms a layer of gas around it thus insulating it and reducing the freezing rate. Freezing hydrated specimens of this size in liquid nitrogen always leads to the development of ice crystals. As they grow, they ruin the original microstructure. Unlike water in hydrated specimens, ethanol does not crystallize when immersed in liquid nitrogen and forms glass-like ice. The frozen prisms are then fractured using a pair of insulated tweezers (they are not available commercially but it is easy to insulate them using short strips of duct insulation) and an insulated scalpel. The 1x1 mm cross section of the specimens allows rapid fixation and impregnation and the 10 mm length makes it possible to obtain several good fragments. The fragments are returned into absolute ethanol to thaw. Finally, they are critical point dried (CPD) and mounted on SEM stubs.

     Specimens such as curd or cheese which contain fat, need to be defatted (the micrograph of Mozzarella cheese above was obtained in that way). The specimens impregnated with absolute ethanol are extracted using three changes of chloroform or n-hexane followed by three changes of absolute ethanol. The specimens are then rapidly frozen, one by one, fractured, and CPD. There is a crucial step when the fragments have to be mounted on the SEM stubs: The fractured plane should be facing up for an easy SEM examination. The fresh plane can be recognized as being shiny in reflected light.

     Why such a complicated procedure? Wouldn't it be easier to examine the surface obtained by cutting the specimen using a blade or a scalpel? No - the milk or soy proteins would be smeared by the blade or scalpel on the cut surface and the images obtained would show artifacts.

Dry-fractured Havarti cheese. The fractured surface is not smooth. Concave globular void spaces indicate where fat globules were located in the fresh cheese.
     Why bother freezing the specimens and fracturing them in liquid nitrogen? Wouldn't it be easier to fracture the specimens after the entire prisms are dried? Oh, yes, it would be easier to do dry-fracturing instead of freeze-fracturing but the images would not be too useful to explain the internal structure of the specimens. Dry fracturing would not proceed along a new flat plane but along the weakest points in the internal structure and the topography of the fractured surface would be very coarse - please check here for information about dry-fracturing. Also check that site for animated ice crystal formation.

January 25, 2005.

It is very difficult if not impossible to show the true structure of a specimen. Most preparatory steps alter that structure. Not all changes may be considered to be artifacts.

     As a microscopist who has spent many years studying the microstructure of yogurt and cheese, I should be able to produce images which would show their true microstructures. They would probably be diagrams - but not micrographs. It is not possible to place a small piece of yogurt or cheese under an electron microscope and to expect an image of the true microstructure.

     As already mentioned, the samples examined by EM must not release gas or vapour. That means that water and other volatile substances must be removed from the sample. This can be accomplished in several ways (including samples in the form of fine suspensions) such as

1. dehydrating and drying the sample and
     1a. examining it by SEM, or
     1b. replicating it with platinum and carbon, and examining the replica by TEM, or
     1b. shadowing it with Pt and examining it by TEM, or
     1c. embedding it in a resin, sectioning it and and examining it by TEM, or
2. freezing the sample, fracturing it while frozen, and
     2a. examining it by cryo-SEM at a very low temperature, or
     2b. replicating it with Pt and C and examining the replica by TEM.
3. The sample can alternatively be examined by so-called environmental scanning electron microscopy (ESEM) which will allow it to retain its aqueous phase.

     None of these procedures can show the structures of yogurt or cheese as they exist in the "fresh" samples. Fresh samples contain water with various substances (including lactose, lactic acid, inorganic and organic salts, milkfat globules, and even bacteria) dissolved and dispersed in it. When we look at a small piece of yogurt, we only see the "wet" surface.

SEM image of yogurt. Left upper corner and right lower corner show the protein matrix of yogurt made by the action of bacteria: streptococci (yellow) and lactobacilli (blue).
     To view the matrix of casein particle chains in yogurt by SEM, the aqueous phase must be removed from the samples. This means, however, that the microstructure of the sample has been changed. Removal of the aqueous phase could lead to a collapse of the structure unless the protein matrix is fixed. That means, however, that a severe change to the original microstructure is introduced. What should be done with the fat globules and droplets of free fat which may be floating in the aqueous phase?
SEM image of Mozzarella cheese. The dark void spaces were occupied by fat globules in the original cheese. Their removal shows the structure of the protein matrix of the cheese and also the distribution of the fat (that had been removed).

     This is a very important question in the case of curd and cheese. Fat cannot be fixed by protein fixatives such as glutaraldehyde, but if it is left unfixed, a part of it will be extracted by liquid carbon dioxide during the critical-point drying procedure and the rest will obscure details of the protein matrix. Researchers often prefer to extract all fat and to study only the protein matrix. If the objective of the study is to retain fat in the protein matrix, the fat must be fixed using an osmium tetroxide fixative which also contains imidazole. However, unless the fats and proteins are in distinctly different forms, it may be very difficult to distinguish them from each other.

     Preparation procedures always alter the microstructure of the samples destined for electron microscopy. Drying for SEM and embedding for TEM remove water, freezing changes the water into ice and may cause ice crystal formation. If the change is intentional, it may allow certain features (e.g., details of the protein structure) to stand out. Such changes are not called "artifacts". This term is reserved for unintentional changes which occur due to negligence or omission.

January 24, 2005.

Following blindly protocols designed for certain biological tissue when working with completely different specimens would not produce images of their true microstructure.

     As the Editor-in-Chief of the scientific journal Food Structure, I reviewed a few manuscripts based on some kind of an idea as if there was a uniform procedure for the fixation of all foods irrespective of the composition, size, and microstructure of the specimens. Common sense suggests, however, that it will take considerably more time for glutaraldehyde or osmium tetroxide to penetrate a compact processed cheese sample than a thin fat-free yogurt sample or that, of two processed cheese samples, the one in the form of a prism with a cross section of 1x1 mm may require 24 hours to get fixed with osmium tetroxide whereas the other, in the form of a 5x5 mm cube, will not be fixed even after 3 or 4 days. "If in doubt, get the answer to your question by an experiment", I used to remind the authors. Osmium tetroxide is particularly useful for that purpose. Leaving 1x1x10 mm prisms of yogurt, low-fat cheese, and high-fat processed cheese in an OsO4 fixative and fracturing a small part of each prism each day will show the penetration of the fixative into the sample as a gradually diminishing light core of the sample's cross section.

     This approach also concerns other procedures such as extraction, dehydration etc. It is true that foods are altered biological tissues - and this fact has to be respected. Ham is much denser than a live or dead muscle, cheese is much denser than yogurt.

January 23, 2005.

Most foods are severely altered (processed) biological tissues. In many of them, the original source is not even noticeable. Example: Wedding cakes do not show that they consist of processed grains (flour), eggs, milk (butter, cream), or sugar cane (sugar) etc. Different bakers would produce cakes of different microstructures. The microscopist must not add to the changes already done to the product.

     My colleague's willingness to instruct his technicians to teach me started a new direction in my research. Since the early seventies I have been using electron microscopes to study the development of structure in milk products until the middle of the nineties. Later on, there was a greater interest in the microorganisms which contribute to the manufacture of milk and many other food products.

     The high resolution of electron microscopy has helped to make advances in medicine and in biology. It was introduced into food science relately late. There was a rapid increase in the number of scientists using electron microscopy in their studies. A journal "Food Microstructure" was established in 1982 and later renamed "Food Structure". It was doing very well, appearing quarterly, but in 1994 it vanished as the publisher, "Scanning Microscopy International" in Chicago (USA) abruptly stopped publishing it without giving reason. The findings of several of my colleagues from those times may be accessed in the "Guest Food Microscopists" sections 1, 2, and 3.

     Commercial production of foods depends on a good knowledge of the raw materials, ingredients, and the finished products manufactured. Experts who modify existing foods and develop new products need to know what affects the microstructure and how microstructure, in turn, controls physical and sensory properties of the food products. Such information may be obtained only by experiments - so let's take some food samples and examine them in an electron microscope.

TEM image of cheese shows a dense microstructure composed of compact proteins (blue), clusters of fat globules (yellow) with the membranes intact. Individual bacteria (red) are dispersed in the aqueous phase (residual whey - white).
     While I searched for information about the appropriate preparation of food samples destined for electron microscopy, I heard opinions of this kind: "Foods are very harshly treated biological tissues. Bread as well as cakes are ground cereals mixed with other ingredients such as sugar, salt, milk or water, partly digested by yeasts and finally damaged by high heat during baking. Meat are muscles of dead animals, already decomposed to a considerable extent. There is no comparison between fresh milk and a piece of aged Cheddar cheese or ground Parmesan cheese. What kind of care do you want to pay to preserve it? That's ludicrous".
    Well, this is some kind of opinion. However, almost any adult and even many children can distinguish between a good and a bad piece of bread, ham, yogurt, cheese etc. Would an electron microscope be helpful to show any important differences? Let's agree that foods indeed are altered biological tissues but let's preserve those changes made on purpose and let's not add additional out-of-control changes.

     There are many conditions to be met if meaningful images (micrographs) of foods are to be obtained from an electron microscope. The specimens placed in a scanning or transmission electron microscope must not release any gases or vapours in vacuo because volatile substances would absorb electrons which are used to magnify the specimens (it is certainly obvious that specimens destined for light microscopy should not release smoke). There is one exception with so-called "environmental scanning electron microscopes" (ESEM) which are constructed and operated to examine hydrated biological samples. These microscopes are relatively new to the field of electron microscopy.

     Preparing the specimens for electron microscopy means retaining their original structure as much as possible and making them resistant to the harsh conditions inside the microscope. Proteins are fixed (denatured under controlled conditions) using specific fixatives, e.g., glutaraldehyde and/or osmium tetroxide, or applying controlled heat-denaturation in a special microwave oven. However, it makes no sense to subject non-proteinaceous specimens such as starches or fats to a similar treatment.

January 21, 2005.

Electron microscopy (EM) showed that gelled milk consists of globules called "casein micelles". Such structures cannot be elastic - unlike structures which consist of fibres. Additional EM studies indicate that the microstructure of foods is related to some of their physical and sensory properties.

     During my young years, children did not have the luxury of having microscopes in their bedrooms and we had to wait until high school to learn about them and try them with a variety of subjects such as onion skin cells, home dust, pollen grains, or yeast cells (Saccharomyces cerevisiae).

     At the Technical University in Brno, I studied technical microscopy as one of the subjects in the first year of chemical engineering. At that time (1949) it was, of course, optical or light microscopy and we examined various fibres such as cotton, wool, silk, etc. Modern polyester fibres only waited yet to be developed and electron microscopes were relatively new high-resolution instruments, too expensive for universities.

     In 1952, I obtained my MEng. degree in chemical engineering, added more experience in food science to what I had learned as a summer student and went back to school (Slovak Academy of Sciences) to study pectic substances in crabapples. In 1957 I started earning for my living for good. The microscopes were forgotten for 15 more years - until the seventies.

     At that time I was a research scientist at a government food research institute in Canada. I was assigned an objective to develop wieners from surplus skim milk powder. It was like in a Czech fairy tale - there the heroes had to find elusive golden-haired princesses, fireball birds, golden-mane horses, or princesses turned into gem stones burried deep at the bottom of the sea. However, each of these heroes had some kind of helper with supra-natural powers - but I even did not have a technician.

     The "hot dog" I made from milk powder looked very fake. Only the pink colour and the spicy smell could probably fool a potential customer. Otherwise it was brittle - not elastic - and instead of being juicy like the comminuted meat product, the milk powder imitation felt like a piece of bad cheese in the mouth. "If I only knew how to look at the microstructure of the two products", I sighed hopelessly, "then, maybe, I would understand, what makes them so different".

     Suddenly it appeared that there was a helper with an almost supra-natural power - the power to show the smallest protein particles in milk. They are called "casein micelles" and their dimensions are between 10 and 20 nanometers (nm). If 1 mm can be divided into 1000 µm, 1 µm can be divided into 1000 nm. The micelles are so small that not even the best optical microscope would show them. It would be as helpless as are our eyes when looking down at bacteria.

     A colleague, an electron microscopist, examined my product and a real "meat-based" wiener. Whereas the wiener consisted of meat fibres, its fake duplicate made from milk powder consisted of tightly packed tiny globules. Therefore, there was no elasticity whatsoever. "Please teach me how to use your electron microscope, so I can investigate these foods in greater detail and learn how their microscopical structure relates to the manner we feel them in the mouth", I asked my colleague.

January 19, 2005.

The books which children read, may affect them for life. Let them read about positive heroes.

     The world never looked the same after the Dutch biologist Antonie van Leeuvenhoek assembled glass lenses and created the first microscope. 
     When I was young (I was born in 1929), scientists like Robert Koch were our heroes - people, who were doing something beneficial for mankind. Dr. Koch was one of "the victors over death" - death caused by bacteria. He saw them, particularly the bacilli of tuberculosis (Mycobacterium tuberculosis), which were then called "Koch's bacilli". At the public school in Czechoslovakia which I attended, we even had theatre performances which showed us how the "bacilli" (represented by other children in white costumes) were dangerous to human health but at the same time were also helpless and unable to harm us if we behaved properly, ate a balanced diet, washed our hands after having used the toilet and before every meal, and never tried smoking.

     There were many additional scientists such as Paul de Kruif and, in particular, Louis Pasteur, who used microscopes for discoveries beneficial for mankind. I find the reading exciting even in these days of high technology here and there and everywhere. Those early scientists had it very difficult, particularly if human prejudice opposed their ideas and experiments. One must admire their resourcefulness.

     Are bacteria indeed so bad that we have to be on guard all the time? Of course, they are, at least a small but im­portant group of them, called "patho­gens". As all microorganisms, they are invisible to a naked eye, so we do not see as to whether our hands are con­taminated, that means whether there are bad bacteria sitting on them. They are the ones which can cause food poisoning or food-borne illness. Just look at some of them at left, enlarged several thousand times by a scanning electron microscope (right: a multisample stage holding 4 bacterial specimens). More recently, bacteria found in various unhygienic situations were examined by SEM.

     |        |  The distance between these two vertical bars is 10 mm (i.e., 1 cm) on the monitor. One millimeter is one tenth of that distance. Divide that 1 mm into one thousand parts and you obtain 1 µm (one micrometer). The bar at the image of the Escherichia coli bacteria shows, how the 1 µm distance at that particular magnification compares to the dimensions of the bacteria. - The figure at right shows the monitor of a scanning electron microscope (Philips XL-30 ESEM) during an examination of a thin layer of bacteria on a bacterial filter. Micrographs obtained by electron microscopy show no colour because they are produced by a beam of electrons. For didactic purposes and appeal they may be colourized digitally.

     Most bacteria are harmless and many are useful to humans (e.g. bifidobacteria, shown at left) by suppressing pathogens. Many other bacteria are used to make a variety of certain foods. Streptococcus thermophilus (shown at right) likes warm milk which it coagulates, with the help of Lactobacillus bulgaricus, into yogurt. Both bacteria belong to a large group of so-called "lactic acid bacteria" (lactobacilli and lactococci). This name has been given to them because they metabolize milk sugar (lactose) in milk and produce lactic acid. This acid coagulates milk proteins and thus produces yogurt, cultured buttermilk, and other fermented milk products. Most cheeses could not be made without bacteria. Lactic acid bacteria including leuconostocs also participate in the production of sauerkraut whereas acetic acid bacteria (Acetobacter aceti) are used to produce "organic" vinegar. Because of their favourable effects on human health, lactic acid bacteria and bifidobacteria are called "probiotic bacteria".

Back to the top


The images are protected by copyright - please contact the author of this website for information.