“Ueber die isolirte Faerbung der Schizomyceten in Schnitt- und Trockenpraeparaten”

The Giant’s Shoulders blog carnival is coming up in two days, and I just realized I still haven’t gotten a post up for it yet. So, here it is.

I put up some quick reviews of several classic microbiology-methods papers for the previous edition of this blog carnival, but didn’t actually get around to putting up the one for what is almost certainly the most well-known microbiology technique: “The Gram Stain”. So, this post is about it:

Gram HC: “Ueber die isolirte Faerbung der Schizomyceten in Schnitt- und Trockenpraeparaten”; Fortschritte der Medicin; 1884; vol 2, pp 185-189

That’s “Regarding the Isolational(?) Coloring of Schizomycetes in Cut- [i.e. tissue sections] and Dried Preparations” in “Medical Progress”. The translation hosted by the American Society for Microbiology uses the word “Differential” where I’ve put “Isolational” – which is probably not quite right either but it’ll have to do for now – but I’ll get to that in a moment.

If you’ve ever been exposed to microbiology labwork before, you’ve almost certainly done or at least watched a procedure referred to as a “Gram stain”. In brief, you smear your sample with bacteria on a glass slide and bake it on, then you dump some purple stuff on it, them some brown stuff, then you rinse it briefly with alcohol, then you dump on some pink stuff, and then rinse it in water and look at it under a microscope. Bacteria that stay the original dark purple-blue color of the original purple/brown stuff are considered “Gram Positive”, and those that don’t instead appear the pink color of the last stain, and are considered “Gram Negative”. Many textbook authors and microbiology instructors will breathlessly proclaim that the Gram Stain reveals two “fundamental” categories of bacteria, but I’ll spare you my rant about that.

Properly speaking, this isn’t actually Gram’s stain, as described in his original paper. The modern variations that we’re all taught in microbiology class were developed later, and I believe they are nowadays based mainly on Victor Burke’s 1922 paper on the subject[1].

Regarding the title of the paper: “schizomycete” is what they used to call most kinds of bacteria. “Mycete” meaning “fungus”, as bacteria were assumed to be “plants without chlorophyll” just like molds and mushrooms, and “Schizo-” meaning “split in two”, since bacteria reproduce by splitting into two cells rather than by producing spores like “other” fungi. I say “most” because things like cyanobacteria (“blue-green algae”) or Green Sulfur Bacteria would have been referred to as “Schizophyta” (“fission-plants”). What Gram was originally trying to do wasn’t to differentiate one kind of bacteria from another, either, but to make it easy to tell bacteria from from the nuclei of cells in bacteria-infected tissue.

For that matter, Gram was really metaphorically standing on the shoulders of Koch and Erhlich, as he was building on their technique for staining “tubercle bacteria” – that is, tuberculosis-causing members of the genus Mycobacterium. Gram mentions that you need to stain this type of bacteria for the “usual” 12-24 hours to make this work, incidentally, as opposed to a few minutes for other “schizomycetes”. This suggests that you are expected to have some idea of what you’re going to find before you use the stain, as opposed to the modern implementation which is supposed to tell you something about what kind of bacteria you’re finding.

Still, Gram does report that some bacteria take the stain and some don’t, giving us a preview of the “differential” character of the modern version. He specifically notes typhoid and some causes of bronchial pneumonia fail to hold the stain. Given that Typhoid Fever is caused by a strain of the “Gram-negative” butt-bacter Salmonella enterica, and there are a number of “Gram negative” bacteria as well as “Gram positive” that can cause pneumonia, this makes sense. He also does mention the use of Bismarck Brown R a.k.a. Vesuvine as a counterstain in order to make the nuclei of the infected cells brown in contrast to the dark blue of the infectious bacteria in the tissue.

For much of the century-and-a-quarter since Gram’s publication, the question of why the Gram stain works was thoroughly investigated, and even today I occasionally hear or read assertions to the effect that the Gram Stain isn’t well understood. I disagree with this just as I think its importance to bacterial identification is grossly overblown, and if you want to know why, I have a previous post all about why the Gram stain works and how we know. You may or may not also be interested in an older post regarding whether or not “acid-fast” bacteria like the ones that cause tuberculosis (which don’t stain at all when using the modern version of the Gram stain) are “Gram Positive” or not. As always, if you spot any errors or have any questions, please let me know…

[1] Burke V: “Notes on the Gram Stain with Description of a New Method.” J Bacteriol. 1922 Mar;7(2):159-82.

“A simplified method of staining endospores”

One more for the “classic papers” challenge:

Schaeffer AB, Fulton MD: “A Simplified Method of Staining Endospores”; 1933; Science; 77; pg 194

If you take a microbiology lab, this is the endospore staining technique (or “technic” as they used to spell it) that you’ll practice. This is a nice, simple, one-page paper. Alice B. Schaeffer and co-author Mac Donald Fulton describe a few of the other variations on endospore staining techniques, then describe how they’ve further simplified what they felt was previously the simplest one, described by a Mr. Wirtz in 1908.

“Endospores” are a sort of “escape pod” for certain specific kinds of bacteria. Unlike spores formed by yeasts and molds, these are not reproductive – each bacterium only produces one thick-coated spore, into which it shoves it’s genetic material and a few vital enzymes to get itself going again later when the spore finds itself in favorable conditions.

Since only a few kinds of bacteria produce these endospores, if you see endospores in your unknown bacterial culture it goes a long way towards helping to identify the bacterial species, so having a simple method for staining your bacteria so that endospores are obvious under a microscope is helpful. (Of course, these days most of us would rather just get a 16s rDNA sequence with PCR, but never mind that for now…)

Endospore stain under a microscope (via Wikipedia)Evidently, Wirtz’s original method involved using Osmium Tetraoxide (“osmic acid”) to stick the bacteria to the slide before staining. Not only is that stuff poisonous, it’s also expensive. I found a site selling sealed glass ampoules containing 1 gram each of this stuff for $35.00 each. Schaeffer and Fulton’s method does away with this in favor of much cheaper and easier heat-fixing (just as is done with the Gram stain and others). They use the dye “Malachite Green” for the initial stain, and steam-heat the dye-covered bacterial slide a few times to sort of “cook” the dye into the thick-walled endospores if they are there. Rinsing then washes the dye out of everything but the endospores, and a light red dye (safranin) is added as a counterstain. The end result is that under the microscope you’ll see light-red bacteria. If any of them form endospores, you’ll be able to see them as smaller green dots – sometimes still bulging inside of bacterial cells, sometimes floating around freely having escaped from the now-empty bacterial cell.

The “Schaeffer-Fulton Endospore Stain” is pretty easy to do, though the occasionally messy steambath part can be annoying. The method is pretty resistant to errors, so it’s not too hard to get good results even if you’ve never done it before.

Incidentally, you can buy Malachite Green at many pet stores – it’s still used as a treatment for “ick” (Ichthyophthirius infestation) in tropical fish.

Hmmm…still a couple of hours before it’s not longer May – Perhaps I can throw in one last post before time’s up…

The Gram Stain Post to End All Gram Stain Posts

Gram stain, Gram stain, Gram stain! Bah. I think it’s time Microbiology grew up and moved out of Medicine’s basement.

Sure, the Gram stain[1] has its uses, but the procedure is grossly over-hyped. “[…]the most important stain in microbiology[…]”[2]! “[…]it is almost essential in identifying an unknown bacterium to know first whether it is Gram-positive or Gram-negative.”![3] “The Gram Stain reaction is an especially useful differentiating characteristic.[…]The Gram reaction turns out to be a property of fundamental importance for classifying bacteria phylogenetically as well as taxonomically.”![4] “[…]differentiates bacteria into two fundamental varieties of cells.”![5] “The Key to Microbiology“![6] [emphasis added…]

Bah! Sure, the Gram stain has its uses, but the hype it gets (even 125 years after its invention) is ridiculous. It’s worse than Harry Potter!

You really want to know what the Gram reaction tells you? Really? Okay, here it is:

A “Gram Positive” reaction tells you that your cells have relatively thick and intact cell walls

A “Gram Negative” reaction tells you that they don’t.

That’s it. That’s about all you can reliably infer from the Gram stain.

Previously, I put up a post describing what was my understanding of the conventional view of why the Gram stain works. Today, I’ll give you a much more detailed – and more correct – explanation of why it works as well as what its real significance is to identification of microbes. But first, a brief one-paragraph rant on why I think the Gram stain has such a hold on microbiology teaching.

I blame the fact that microbiology education is still largely in the shadow of medical technology education. When you artificially exclude the 99+% of organisms that aren’t associated with human diseases, the tiny number left do, indeed, seem to largely separate into two phylogenetic categories. Judging by what I’ve encountered thus far, it seems you get a lot of Proteobacteria (especially ?-Proteobacteria, like E.coli), which are “Gram-negative”. You also get a lot of Firmicutes (Bacillus, Streptococcus, Staphylococcus, etc.), and a couple of scattered Actinobacteria (Mycobacterium, for tuberculosis and leprosy, Corynebacterium for diptheria…). Both of these are considered “Gram-positive” (although if you use the standard procedure these days, the Mycobacteria may show no reaction at all). That’s, what, 3 phyla out of about 25 eubacterial and archael phyla? If we throw in Syphilis and Chlamydia, that’s still only 20% or so of the currently recognized prokaryotic phyla. If your microbiology classes assume everybody is training to be a medical technologist or clinical microbiologist, then the Gram stain becomes inflated in importance.

Enough of that – here’s a quick review of how the Gram stain works. Solutions of “Crystal Violet” (a purple dye) and Iodine are applied to cells fixed to a slide, where they soak in and precipitate in the cells. A “decolorizer” (usually ethanol) is applied to see if it will wash this dye precipitate out of the cells. A different, lighter-colored dye (such as safranin) is added so that the cells which DO have their dye washed out can be seen as well. In the end, “Gram positive” cells are a dark purple from the crystal violet/iodine that was not washed away, and “Gram negative” cells are not dark purple. (Usually they are pink, from the safranin, assuming that’s the dye used as the counterstain.)

Note that this does not differentiate cells into “two fundamental types” as is often claimed. You actually get four types: Groups of cells that are normally always “Gram positive”, Groups of cells that are normally always “Gram negative”, Groups of cells that are normally sometimes “Gram positive” and sometimes “Gram negative” (“Indeterminate”, or as I like to call it, “Gram-biguous”), and groups of cells that are normally NEITHER Gram-positive nor Gram-negative, like Mycoplasma, which aren’t dyed at all by the process. Incidentally, phylogenetically speaking, Mycoplasma is one of the “Gram positive” Firmicutes, just like Bacillus and Staphylococcus.

It’s kind of interesting to me that the Gram stain reaction has been such a mystery up until a century after its invention. What is it that makes “Gram positive” cells retain the dye while “Gram negative” ones don’t? Along the way, it seems like nearly every part of the bacterial cell was hypothesized to be the reason for the Gram reaction – lipids, carbohydrates, nucleic acids, “Magnesium ribonucleates”, and so forth. Davies et al, 1983, includes a table listing many of these and referencing historical papers making the claims. The fact that the reaction had something to do with the cell wall seems to go back quite a while, though the “Magnesium ribonucleates” idea doesn’t seem to have been entirely abandoned until the mid-1960’s[7]. It was also hypothesized that the “Gram positive” cells simply absorb more dye and therefore take longer to “decolorize”.

It turns out that “Gram-positive” cells actually don’t, necessarily, take up more dye than Gram negative ones. This was tested by taking a set concentration of bacterial cells and adding them to a set concentration of dye. After letting them soak, the samples were centrifuged to remove the bacteria, and the amount of dye found to be missing from the liquid was taken as the amount absorbed by the cells. They found that some Gram negative cells actually took up more dye than the Gram positives did. So much for that idea.[8]

Even relatively recently, I’ve seen it written that the bacterial cell wall, specifically, is what holds onto the stain, but even that turns out not to be true. Although the cell wall is the structure that seems to be responsible for the Gram reaction, in the late 1950’s it was demonstrated that it was not actually the staining of the cell wall that caused the reaction, but rather the ability of the cell wall to keep the decolorizer out of the cell.[9]

Apparently, the Crystal Violet/Iodine complex itself doesn’t even play a vital role. The complex apparently dissolves again more or less instantly as soon as the decolorizer touches it[10], and it’s even possible to differentiate “Gram positive” and “Gram negative” with simple stains like methylene blue or malachite green, if you’re clever about it[11]. The latter authors set up a clever test with crushed cell material, dye, and paper chromatography. They had the decolorizer soak into the paper, past a spot where dye-soaked cell material from Gram-positive and Gram-negative cells was placed, and watched for obvious differences in the amount of time it took the dye to be carried out by the decolorizer. Incidentally, my quick examination of this paper makes it look like cheaper 100% isopropyl alcohol (“rubbing alcohol”) might be slightly better than the standard 95% ethanol for Gram stains.


So, here we are at 1970 or so, and we already know that the Gram reaction is entirely based on how well the cell wall structure prevents organic solvents (like ethanol) from soaking into the cell to dissolve the dye complex. Yes, the mystery of why the Gram stain works in normal cells was largely solved by the Nixon era.
A few corners of the mystery remained, though. Why do “old” cultures of “Gram positive” cells often end up staining “Gram negative”, for example? Why do some kinds of cells seem to be sometimes Gram positive and sometimes Gram negative in the same culture? What, exactly, is really happening to the cell, deep down, during the staining process?

In 1983, the Gram Stain made the great technological leap into the 1930’s, when a variation of the technique was devised which allowed the Gram Stain to be observed by electron microscopy[12]. Using a funky platinum compound in place of iodine, the electron microscope reveals exactly where the dye complex is at any particular stage of the Gram stain process. Using this technique, it was possible to see how the decolorizer disrupts the outer membrane of classically-Gram-negative organisms and to see that the decolorizer potentially damages the cell wall and interior membrane, possibly allowing cell material to leak out (or decolorizer to get in and dissolve the dye complex). It was also seen that the dye complex permeates the entire cell, not just the cell wall.[13]

If you’ve been wondering about the sometimes-Gram-positive-sometimes-Gram-negative cells, the same technique was also used to investigate this. As suspected, it turns out that the “old cultures become Gram negative” problem is due to the cell walls breaking down as the culture ages. Bacteria are continuously, simultaneously, building up and tearing down their cell walls, in order to be able to grow and divide. As nutrients run out, the bacteria run out of material to rebuild cell walls, while the cell-wall degrading enzymes keep on chugging. Breaks in the cell wall occur, and through these breaks the decolorizer can get in and rapidly dissolve the dye. Actinobacteria can have a similar problem, but rather than only being in “old” cultures, apparently weaknesses appear briefly during cell division, and if a particular cell happens to be at this stage of growth when you stick it on a slide, heat-fix, and Gram stain it, the weakness at the septum where the division is occuring can crack and allow the decolorizer in, resulting in a “Gram negative” response even while surrounding cells of the same kind might still be “Gram positive”.[14]

This brings us to archaea and some eukaryotes (i.e. yeasts). Yeasts stain “Gram positive” normally. Although their cell walls are completely different chemically than bacterial cell walls, they are quite thick (microbially speaking). Poor, neglected Archaea seem to be all over the place in terms of Gram reaction. Since their Gram reaction doesn’t tend to correlate to any particular phylogenetic grouping[15], it seems nobody really pays much attention to their Gram stain reaction. On the other hand, and on the subject of “Gram-biguity”, I thought the investigation of Methanospirillum hungatei[16] was interesting. M.hungatei is an archaen that grows in chains. When Gram-stained, the cells on the ends of the chains are “Gram positive”, while the others have no Gram reaction at all. It turns out that the chains are covered by a sheath, and the only contact with the outside world is through thick “plugs” in the cells at the ends of the chains. These “plugs” act like thick cell walls, allowing the Gram stain dye material to soak in but excluding the decolorizer, while the sheath keeps the rest of the cells from soaking up any stain at all.

There you have it – a relatively detailed history and explanation for the Gram stain, and you didn’t even have to get through some obnoxious paywall to read it. Aren’t you lucky?

Comments, suggestions, and corrections, as always, are welcome.

[1] Gram, HC.”Ueber die isolirte Faerbung der Schizomyceten in Schnitt-und Trockenpraeparaten.” Fortschitte der Medicin. 1884 Vol. 2, pp 185-189.

[2] Popescu A, Doyle RJ. “The Gram stain after more than a century.” Biotech Histochem. 1996 May;71(3):145-51.

[3] Brock TD, Madigan MT, Martinko JM, Parker J. “Biology of Microorganisms (7th Edition).” 1994. Prentice Hall, Englewood Cliffs, NJ pg. 46

[4] ibid, pg. 715

[5] Beveridge TJ.”Use of the gram stain in microbiology.” Biotech Histochem. 2001 May;76(3):111-8.

[6] McClelland, Rosemary. “Gram’s stain: The key to microbiology – isolate identification method – Tutorial” Retrieved 20070810 from http://findarticles.com/p/articles/mi_m3230/is_4_33/ai_74268506/print

[7] Normore WM, Umbreit WW.”Ribonucleates and the Gram stain.” J Bacteriol. 1965 Nov;90(5):1500.


[9] BARTHOLOMEW JW, FINKELSTEIN H.”Relationship of cell wall staining to gram differentiation.” J Bacteriol. 1958 Jan;75(1):77-84.


[11] Bartholomew JW, Cromwell T, Gan R.”Analysis of the Mechanism of Gram Differentiation by Use of a Filter-Paper
Chromatographic Technique.” J Bacteriol. 1965 Sep;90(3):766-77.

[12] Davies JA, Anderson GK, Beveridge TJ, Clark HC.”Chemical mechanism of the Gram stain and synthesis of a new electron-opaque marker for electron microscopy which replaces the iodine mordant of the stain.” J Bacteriol. 1983 Nov;156(2):837-45.

[13] Beveridge TJ, Davies JA.”Cellular responses of Bacillus subtilis and Escherichia coli to the Gram stain.” J Bacteriol. 1983 Nov;156(2):846-58.

[14] Beveridge TJ. “Mechanism of Gram Variability in Select Bacteria.” J Bacteriol. 1990 Mar;172(3):1609-20.

[15] Beveridge TJ, Schultze-Lam S. “The response of selected members of the archaea to the gram stain.” Microbiology. 1996 Oct;142 ( Pt 10):2887-95. (Abstract)

[16] Beveridge TJ, Sprott GD, Whippey P. “Ultrastructure, inferred porosity, and gram-staining character of Methanospirillum hungatei filament termini describe a unique cell permeability for this archaeobacterium.” J Bacteriol. 1991 Jan;173(1):130-40.

Are “Acid-Fast” bacteria Gram-positive or Gram-negative?

I was wondering what today’s post ought to be – but a Google™ search that reached the page posed an interesting question and made it easy.

Someone from a Miami, Florida college got to this site after asking Google:
“Assuming you could stain any cell, would an acid-fast [bacterium] be gram-positive or gram-negative?”

Therefore, today’s post will deal with some Microbial Physiology.

The easy answer is, of course, “no“.

A more useful answer, though, is that it depends on what you mean by “Gram-positive”.

If one takes the “Assuming you could stain any cell” part of the question to mean that you’ve done whatever it takes to get the Crystal Violet/Iodine into the cell wall, and you mean “will the cell still look purple instead of pink at the end of the Gram Stain process”, then I’m pretty sure the answer would be yes, that it would be “Gram Positive”. It actually IS possible to stain “acid-fast” bacteria with the Gram stain. The catch is that it takes 12-24 hours of staining (according to Gram’s original paper) rather than a minute or so. This still counts as Gram-positive, though, and in fact the whole Phylum of Actinobacteria (including the Mycobacterium genus) is considered “High G+C Gram-positive”. (If the query was for an exam or homework problem, this is probably the answer you’re looking for.)

On the other hand, if you mean “does the cell wall structure of an acid-fast bacterium better resemble Gram-positive or Gram-negative bacteria?”, you can make an argument that instead of a nice simple “inner membrane surrounded by a thick peptidoglycan-layer cell wall”, the “acid-fast” bacterial cell wall looks more like a complex gram-negative-type cell wall, having multiple layers, with special proteins that form channels through them to let substances in and out of the cell through the otherwise penetration-resistant outer layer, just as Gram-negative bacteria have through their outer membrane. (On yet another hand, those outer layers are tough like a gram-positive bacterium rather than fragile like a gram-negative bacterium’s outer membrane, and don’t dissolve in alcohol.)

Therefore, if you’re speaking in terms of taxonomy, and by “Gram-positive” you mean firmicutes which, as far as I know at the moment, are really the only class with the simple, officially-gram-positive-type cell wall structures and therefore are usually what is meant when someone says “Gram-positive” (someone please correct me if I’m wrong here), the obvious differences with “acid-fast” type cell walls can at least make a good argument that they are “not ‘Gram Positive’ bacteria”.

But if you put that on your homework or exam answers, don’t blame me if you get marked wrong…

I actually found a pretty nice illustration on the University of Capetown website showing the differences in cell wall structures here, which might help.

So, to summarize – Officially, they’re either “neither” or “Gram-positive”. Unofficially, you can probably argue either way. Hmmm. I should try to work in a post on bacterial taxonomy one of these days.