The Gram Stain: Its Per­sis­tence and Its Quirks

by Elio

What is more em­blem­atic of our sci­ence than the Gram stain? Since its in­ven­tion 130 years ago, it has been in fre­quent and con­tin­u­ous use. It con­ve­niently places most bac­te­ria into one of two groups, the Gram-pos­i­tives or the Gram-neg­a­tives. Gram stain­ing is cheap, ef­fec­tive, quick, and rel­a­tively easy to in­ter­pret. Its most use­ful ap­pli­ca­tion is in the clin­i­cal set­ting. When ex­am­in­ing a smear of, say, pus from an ab­scess, this stain of­ten al­lows to in­clude for con­sid­er­a­tion roughly half the clin­i­cally rel­e­vant bac­te­r­ial species while ex­clud­ing the oth­ers. Or con­sider a pa­tient with menin­gi­tis. Here, speed is of the essence be­cause treat­ment must be ini­ti­ated right away. A Gram stain of the spinal fluid may re­veal within min­utes the pres­ence of Gram-pos­i­tive cocci, (prob­a­bly pneu­mo­cocci), Gram-neg­a­tive cocci (al­most cer­tainly meningo­cocci), or Gram-neg­a­tive slen­der rods (most likely Haemophilus in­fluen­zae). This can make a crit­i­cal dif­fer­ence in the choice of an­tibi­otics that have to be ad­min­is­tered in great haste. How­ever, when it comes to char­ac­ter­iz­ing the bac­te­ria in an en­vi­ron­ment, its use­ful­ness di­min­ishes, in part be­cause it is not al­ways in step with tax­on­omy, which I'll dis­cuss be­low. And there yet is an­other side to this story. Quite a few bac­te­r­ial species that stain pos­i­tive early in the growth of a cul­ture be­come Gram-neg­a­tive later on. Does this de­tract from the value of Gram's method? It may, but not in the hands of a per­son ex­pe­ri­enced in its use.

Hans Chris­t­ian Joachim Gram. His pa­per on the Gram stain pub­lished in Fortschritte der Med­i­cin, 2(6), 185–189 (1884), was en­ti­tled: "Ue­ber die isolirte Fär­bung der Schizomyceten in Schnitt- und Trocken­prä­paraten" (roughly: "On the dif­fer­en­tial stain­ing of Schizomycetes in dry prepa­ra­tions and sec­tions." Schizomycetes, or di­vid­ing fungi, was a com­mon term for bac­te­ria in the early days of mi­cro­bi­ol­ogy. Source

A Bit of His­tory: How to Stain Bac­te­ria in Tis­sue Slices

In the early days of Bac­te­ri­ol­ogy, stain­ing was a big deal (cul­tur­ing was just com­ing into its own). Stain­ing pro­ce­dures were aided by the avail­abil­ity of strong dyes, the syn­thetic ani­line dyes, that were first made around the mid­dle of the 19th cen­tury. Stain­ing bac­te­ria on a smear with these dyes worked fine, but vi­su­al­iz­ing them in tis­sue slices—which is one way to ful­fill Koch's first postulate—was an­other mat­ter. Here, all sorts of cel­lu­lar and ex­tra­cel­lu­lar ma­te­r­ial also take up the dyes used, which makes it nearly im­pos­si­ble to dis­tin­guish the small bac­te­ria. The prob­lem, it was re­al­ized, could be solved by us­ing ways to se­lec­tively stain the bac­te­ria. One way to do it could be to de­col­orize the other stuff. The first suc­cess­ful tech­nique based on this prin­ci­ple used acid to dif­fer­en­tially re­move the stain. For­tu­itously, this method was de­vel­oped to dis­tin­guish tu­ber­cle bacilli, which uniquely re­tain a dye af­ter acid treat­ment (hence they are "acid fast"). Other bac­te­ria are bleached by acid along with what's in the rest of the sam­ple.

Find­ing ways to dif­fer­en­tially stain other bac­te­ria in tis­sues re­quired a lot of work and all sorts of tri­als. One strat­egy that clearly came to mind was to use mor­dants, chem­i­cals that help "set" a dye. This was stan­dard prac­tice in the tex­tile in­dus­try, the prin­ci­pal user of these syn­thetic dyes and the de­vel­oper of mor­dants for in­ten­si­fy­ing the col­ors. Many pos­si­ble mor­dants were used to dye bac­te­ria as well, e.g., phe­nol in the acid-fast stain­ing pro­ce­dure. These meth­ods were known to the Dan­ish physi­cian H. C. Gram (1853–1938) who wanted to see the bac­te­ria in slices of lung tis­sue from pa­tients who had died of pneu­mo­nia. He did not have in mind dis­tin­guish­ing one kind of bac­terium from an­other, the main pur­pose to­day for the pro­ce­dure that bears his name.

Gram tried as a mor­dant a com­mon io­dine so­lu­tion, Lugol's, which he al­most surely had within arm's reach. When us­ing a vi­o­let stain, he soon found out that adding the io­dine so­lu­tion pre­cip­i­tated the dye. In other words, it made a mess. Hop­ing to clean up the prepa­ra­tion, Gram tried flood­ing it with ethanol. He found that, if he did it care­fully, the tis­sue com­po­nents were de­col­orized whereas some bac­te­ria re­tained the vi­o­let dye and oth­ers did not. The col­or­less ones were now hard to see, but adding a "coun­ter­stain" of a dif­fer­ent color (red) made them read­ily vis­i­ble. So, Gram-pos­i­tives be­come "pur­ple," Gram-neg­a­tives "red." Be­ing a mod­est man, Gram stated: I have there­fore pub­lished the method, al­though I am aware that as yet it is very de­fec­tive and im­per­fect; but it is hoped that also in the hands of other in­ves­ti­ga­tors it will turn out to be use­ful. How true is the last state­ment! It bears stress­ing that Gram's dis­cov­ery was any­thing but ac­ci­den­tal, de­spite what has been stated in some text­books.

The steps in Gram stain­ing. If you're go­ing to do it, con­sider wear­ing gloves or else you may find out that your fin­gers are vis­i­bly Gram-pos­i­tive. Source

How Does It Work?

The crux of the Gram stain is the ex­po­sure to al­co­hol (or to the other or­ganic sol­vents that are of­ten used) dur­ing de­col­oriza­tion. For the Gram-neg­a­tives, most of which pos­sess a thin pep­ti­do­gly­can cell wall, this dam­ages the cy­to­plas­mic mem­brane and al­lows the es­cape of the dye-io­dine com­plex. The Gram-pos­i­tives, on the other hand, typ­i­cally have a thick pep­ti­do­gly­can layer that some­how im­pedes the pas­sage of al­co­hol or other sol­vents. It is thought that per­haps sol­vents ren­der the en­ve­lope im­per­me­able by de­hy­drat­ing the peptidoglycan—a good thought that ap­pears in some writ­ings but for which I have not found the pri­mary sources. I would ap­pre­ci­ate it if some­one could point me to them.

One fur­ther thought. Al­most uni­ver­sally, the Gram stain is car­ried out on heat-fixed smears on a mi­cro­scope slide. This time-tried step is of course sim­ple and con­ve­nient. How­ever, heat and dry­ing may fur­ther weaken the cell en­ve­lope com­po­nents, thus adding to the loss of Gram-pos­i­tiv­ity. For the sake of gain­ing fur­ther un­der­stand­ing of the mech­a­nism of the Gram stain, per­haps stud­ies should be car­ried out on un­fixed cells. I ex­pect that un­der those con­di­tions some cells may not take up the dye or the io­dine so­lu­tion, but this could prob­a­bly be tam­pered with. Any­body know of such stud­ies?

The Gram Stain En­coun­ters Tax­on­omy

In a sim­pler world, all bac­te­ria with a sin­gle mem­brane, the mon­o­derms, would be Gram-pos­i­tive, all those with two (the di­derms), Gram-neg­a­tive. Put tax­o­nom­i­cally, all Fir­mi­cutes (e.g., Bacil­lus, Clostridia, Staphy­lo­coc­cus, Strep­to­coc­cus) and Acti­nobac­te­ria (e.g., Corynebac­terium, My­cobac­terium, Strep­to­myces) would be Gram-pos­i­tive, all the rest (e.g., Es­cherichia, Haemophilus, Pseudomonas, Neis­se­ria), Gram-neg­a­tive. Hah! Were it that sim­ple! Each of these groups in­cludes out­liers that stain op­po­site from the ex­pected way. An il­lu­mi­nat­ing es­say on evo­lu­tion­ary con­sid­er­a­tions that helps un­der­stand this state of af­fairs was re­cently writ­ten by Jeff Erring­ton.

Most mon­o­derms fall con­ve­niently in the Gram-pos­i­tive camp all right, but there are im­por­tant ex­cep­tions. First, the Fir­mi­cutes (so called for the "firm­ness" of their thick pep­ti­do­gly­can cell wall) en­com­pass the wall-less my­coplas­mas and their al­lies, bac­te­ria that can­not be stained by the Gram stain at all, which makes the point moot here. Next, some typ­i­cal Fir­mi­cutes (the Neg­a­tivi­cutes) stain Gram-neg­a­tive. To make things worse, at least one fir­mi­cute, Ace­tonema, ap­pears to have the pro­to­typ­i­cal Gram-neg­a­tive prop­erty, an outer mem­brane (a fas­ci­nat­ing story for some other time, but the rea­son I got into this sub­ject in the first place). Hav­ing a thin pep­ti­do­gly­can layer may place whole phyla (e.g., the green non-sul­fur bac­te­ria or Chlo­roflexi) within the Gram-neg­a­tives, de­spite be­ing mon­o­derms.

Left: Thin sec­tion of a My­cobac­terium phlei cell con­tain­ing TPt-crys­tal vi­o­let com­plex (darkly stained re­gions in cy­to­plasm). TPt is trichloro(eta 2‑ethylene)-platinum(II). Right: Dots of TPt-crys­tal vi­o­let com­plex are re­vealed by EDS (En­ergy-dis­per­sive X‑ray spec­troscopy) in a cell sim­i­lar to the one on the left. The dots are evenly dis­trib­uted within the cy­to­plasm. Bar = 200 nm. Source

When it comes to con­sis­tency, the di­derms fare no bet­ter. Some, such as Deinococ­cus, stain Gram-pos­i­tive be­cause they have an un­usu­ally thick pep­ti­do­gly­can layer, de­spite be­ing di­derms and pos­sess­ing the two mem­branes typ­i­cal of Gram-neg­a­tives. And if this were not bad enough, le­sions that af­fect cell in­tegrity can ren­der oth­er­wise Gram-pos­i­tives to stain as Gram-neg­a­tives. Au­then­tic Gram-pos­i­tives such as mem­bers of the genus Bacil­lus will grad­u­ally lose their Gram-pos­i­tiv­ity and some of their pep­ti­do­gly­can cell wall af­ter they stop grow­ing. Note that I have not men­tioned the Ar­chaea, which are mainly Gram-neg­a­tive, with some in­ter­est­ing ex­cep­tions.

Gram Vari­abil­ity

Clin­i­cal mi­cro­bi­ol­o­gists are used to liv­ing with "Gram in­de­ter­mi­nate" bac­te­ria. The mat­ter of Gram vari­abil­ity is ac­tu­ally quite a bit more in­tri­cate than I made it sound. The late Terry Bev­eridge, whose stel­lar con­tri­bu­tions to mi­cro­bial cell struc­ture were cel­e­brated in these pages, stud­ied this phe­nom­e­non in de­tail, us­ing a va­ri­ety of or­gan­isms. He clev­erly syn­the­sized an io­dine ana­log, trichloro(eta 2‑ethylene)-platinum(II), that is elec­tron scat­ter­ing, hence could be used to look at sec­tions of Gram-stained bac­te­ria in the elec­tron mi­cro­scope. Read his pa­per for de­tails. In brief, any­thing that dam­ages or weak­ens the cell en­ve­lope may turn Gram-pos­i­tives neg­a­tive. This in­cludes the thin­ning of the cell wall pep­ti­do­gly­can layer that hap­pens as some cul­tures age, or the weak­en­ing of the pro­tein S‑layer, with which some bac­te­ria and many ar­chaea are en­dowed. Gram stain­ing im­poses great stress on the in­tegrity of such or­gan­isms, so it's not sur­pris­ing that Gram-pos­i­tiv­ity can be read­ily lost. Note that it's rel­a­tively easy for a Gram-pos­i­tive to be­come Gram-neg­a­tive whereas the con­verse is not the case. This may not be the whole story be­cause frag­ments of Gram-pos­i­tives have been re­ported to stain pos­i­tive, pos­si­bly sug­gest­ing an al­ter­na­tive mech­a­nism for the Gram stain.

Terry Bev­eridge, 1945−2007. Source

Sum­mary

Let me wrap this up by say­ing again that most mon­o­derms stain Gram-pos­i­tive, most di­derms, Gram-neg­a­tive. How­ever, there are im­por­tant ex­cep­tions. Know­ing that an or­gan­ism is a mono- or a di­derm doesn't tell you for sure what its Gram stain­ing prop­er­ties will be. Some mon­o­derms stain neg­a­tive, some di­derms, pos­i­tive. What seems to count is how thick and in­tact the pep­ti­do­gly­can layer is. And, I should add, the Gram stain is not use­ful to dis­tin­guish the Ar­chaea, which don't have pep­ti­do­gly­can and stain in a seem­ingly un­pre­dictable man­ner. It's a mir­a­cle that de­spite these shenani­gans, the Gram stain re­mains a ma­jor tool in the clin­i­cal mi­cro­bi­ol­ogy lab. And, for all the ex­cep­tions, it proved to be a key tool in the elu­ci­da­tion of cen­tral prop­er­ties of the bac­te­r­ial cell.

I thank Jeff Karr, the ASM archivist, for sup­ply­ing orig­i­nal ma­te­r­ial for this pa­per.