Upsala J Med Sci 105: 3 140,2000 

Morphologic Conversion of Helicobacter Pylori From Spiral 
to Coccoid Form 

Scanning (SEM) and transmission electron microscopy (TEM) 
suggest viability 

Roger WillCn,' Birgitta CarlCn,2 Xin Wang," Nicos Papadogiannakis? Rolf Odselius3 and Torkel Wadstrom4 
'Department of Pathology, Sahlgrenska University Hospital S-413 45 Goteborg, Sweden, 2Department of Pathology, 

'Electron Microscopy and 4Clinical Microbiology Lund University Hospital and 5Departnzent of Pathology, 
Huddinge University Hospital 

ABSTRACT 

Helicobacter pylon is a pathogen associated with type B gastritis, peptic ulcer disease, gastric 

atrophy, and stomach cancer. H. pylori exists in two morphological forms, spirals and coccoids. 

The latter has been described as viable but non-cultivable. The role of the coccoid form in the 

pathogenesis of gastric disease is disputed. Some authors consider the coccoid form to be a 

degenerative or dead form of H.pylori, while others consider it a resting but still metabolically 

active form. 

This study reports the conversion fiom spiral to coccoid form ultrastructurally. Dense material is 

accumulated in the periplasmic space, the spiral bacteria bend and the outer membrane is separated 

from the inner cell wall layer. Remodeling of inner structures takes place, ending in the coccoid 

form of the bacteria with preserved light polyphosphate areas. Reduction of surface takes place by 

production of surface membrane vesicles, which later are squeezed off. The finding of preserved 

subcellular structures +d intact double membranes in combination with degenerative forms 

suggests that some of the coccoids are viable. Scanning electron microscopy (SEM) demonstrates 

coccoid form of bacteria with slightly ruffled surfaces but no spiral forms. 

INTRODUCTION 

Helicobacter pylon is a human pathogen associated with type B gastritis, peptic ulcer disease and 

gastric atrophy and stomach cancer [32]. An oral challenge with H.pylori caused an infection in 

human volunteers, monkeys, gnotobiotic pigs, nude and euthymic mice inducing histopathological 

changes similar to natural Type B gastritis in man [32]. 

31 



H. pylori exists in two morphological forms, spirals and coccoids. The latter have been described as 

"viable but non-cultivable"(VBNC). The coccoid form of H. pylori may be viable and can revert to 

cultivable forms in mice [7, 321 but are no longer cultivable on conventional media [3, 5, 9, 10, 16, 

301. The role of the coccoid form in pathogenesis of H. pylori-associated gastritis has been disputed. 

Some authors consider the coccoid form as a degenerative or dead form of H. pylori [ 16,21,25], 

while others consider it as a resting but still metabolic active form [2, 3, 12,22, 29, 321. The aim of 

this study was to demonstrate the conversion of spiral form of H. pylon to coccoid forms, and to 

show both degenerative and intact subcellar morphology suggesting that some of the coccoid 

bacteria may be viable. 

MATERIAL AND METHODS 

Bacterial strains and culture 

H. pylori strain 553193 was a freshly isolate from human gastric biopsies. The strain was grown on 

GAB-CAMP agar supplemented with horse serum, 10% and incubated for 48 h at 37" C under 

microaerophilic conditions to obtain a high yield of spiral shaped H.pylori [32]. To obtain viable 

but non-cultivable H.pylori bacteria, 3-5 day-old agar cultures were harvested and resuspended 

in 20 ml of Ham's F12 medium supplemented with 10% calf serum (Flow laboratories, Irvine), and 

incubated in a micro-aerobic environment for 3 days at 37" C and then stored at 4°C. If no growth 

was observed aRer incubation for 5 days on GAB-CAMP agar at 37"C, harvested cells were 

defined as viable but non-cultivable [32].All bacteria were analyzed for the presence of spiral forms 

in the coccoid suspension. No spiral forms were found. 

Transmission electron m i c r o s c o p y m )  

During the spiral to coccoid conversion period samples were taken and immediately fixed in 

2%glutaraldehyde, in 0.1 M sodium cacodylate buffer (pH 7.2), postfixed in 2% osmium tetroxid, 

in S-colloidine buffer (pH 7.2), dehydrated in graded ethanol and embedded in agar resin 100. 

Semithin section was cut and examined by light-microscopy. A representative area was chosen and 

ultra-thin sections, 50 nm were cut on a Reichert-Jung Super Nova Ultramicrotome and 

contrasted with uranyl acetate and lead citrate. The grids were examined in a Philips CM 10 

transmission electron microscope at 60 kV. 

32 



Scanning electron microscopy ( S E M )  

Coccoid Helicobacter cells were placed i n  a fixative solution containing 2.5%(wt/vol.) 

glutaraldehyde for 24 h at 22" C. The preparations were rinsed 3 times in Soerensen's phosphate 

buffer, and then dehydrated in a graded alcohol series ending with 99% ethanol containing a 

molecular sieve, and then critical point-dried in liquid carbon dioxide. The specimens were 

mounted on metal stubs and sputtered in a Polaron E 5400 sputter coated with gold for SEM. 

Each specimen was studied in a Philips SEM 515 unit at a magnification of x 312 to x 

5700.Photographs of representative findings were obtained. 

RESULTS 

SEM showed rounded coccoid H. pylori after 5 days of cultivation on GAB-CAMP agar. No 

spiral shaped H. pylori were detected. The coccoids were seen in clusters and some debris was 

noted (Fig 1). 

Fig 1. Clusters of Coccoid bacteria 
after 5 days cultivation on GAB-CAMP 
agar with rounded contours and some 
debris. No spiral forms are detected. 
SEM x 5,700 (original magnification). 

In TEM transformation of the spiral form of H.pylori to coccoid form occurred via an 

intermediate U-shaped form. It is initiated by ingrowth of the periplasmic space on one side of 

the bacteria separating the double membrane into two membranes (Fig 2).Accumulation of cell 

dense material could be seen. Later the U-shaped form became predominant with enhancement 

of electron dense material (Fig 3). Transection of such U-bent bacteria yielded two forks with a 

light space in-between. The membranes were separated into an inner and outer part, where the 

forks were either covered by a single layer at the center and double layer at the periphery (Fig 4). 

Sometimes at this development stage a part of the spiral form was clearly visible while the other 

part became more diffuse (Fig 5 ) .  

3- 1001 26 33 



Fig 2 Fig 3 

Fig 2.H.pylori spiral form with separation of inner and outer membrane by the periplasmic space 
with accumulation of dense granular material (Arrow). TEM 105,0OO(original magnification). 

Fig 3.U-formed spiral bacteria. Note cytoplasmic vesicles, polar density and light areas, 
supposedly polyphosphate-rich areas. Diameter 1.34 pm. TEM x 21,000 (original magnification). 

Fig 4 Fig 5 

Fig 4.Transection of U-bent H.pylori spiral form with a clear space between forks. Note remains 
of inner membranes around light and dense structures. TEM x 52000 (original magnification). 

Fig 5.Partly formed coccoid H.pylori. Condensation of dense material in one part of the bacteria, 
double membranes and shedding of cytoplasmic vesicles. TEM x 52000 (original magnification). 

34 



Fig 6 Fig 7 

Fig 6.Nearly finished H.pylori coccoid. Note light areas and cytoplasmic vesicles sometimes 
containing dense material. TEM x 52,000 (original magnification). 

Fig 7.Fully developed H. pylori coccoid. Note double membranes, dense and light areas and part 
of a flagella (arrow). TEM x 105,000 (original magnification). 

Fig 8 Fig 9 

Fig 8.Mixture of transected spiral H. pylori (arrow) (Diameter 0.41-0.45 pm) and coccoid forms 
(arrow head) (1.04-1.16 pm). TEM x 15,500 (original magnification). Fig 9.Coccoid H. pylori 
with morphologically clear degeneration and fragmentation of membranes and central dense 
area. TEM ~73,000 (original magnification). 

35 



The electron dense material became gradually more rounded in the center of the coccoid-like 

structure. The center also contained light spaces, supposedly holding phosphate-rich material. 

During the transformation many cytoplasmic vesicles were produced, sometimes squeezed in 

the space outside the coccoid bacteria (Fig 6). The transformation strongly suggests an 

dynamic process. The double membrane system remained but periplasmic dense material 

disappeared subsequently (Fig 7). Sometimes flagella-like structures were seen in connection 

with the fd!y developed coccoid form (Fig 7). 

While transection of the spiral form was between 0.4-0. 8 pm, the coccoid form demonstrated 

a diameter of 0.9-1.4 pm (Fig 8). A progressive loosening and detachment of the outer 

membrane vesicles accompanied this transformation. Degenerated coccoid bacteria were 

also found with fragmentation of membranes and disruption of centrally dense material (Fig 9 ). 

DISCUSSION 

The mode of transmission of H.pylori from man to man is still unclear. Two theories exist: 

one oral-oral and one oral-fecal transmission through contaminated water [6, 1 1, 171. Lately 

also aniatrogenic spread through the gastroscope and vectors like cats and houseflies has been 

suggested [6]. The spiral form is very sensitive for air drying and therefore the coccoid form of 

H.pylori has been suggested to represent the transmission form. Therefore much efforts have 

been put into its characterization. The coccoid form is non-cultivable on conventional media 

and accordingly pro and cons regarding viability have been put forward. The problem of non- 

cultivable but in water viable coccoid forms as Vibrio, Shigella, Legionella, Campylobacter and 

Escherichia coli are well known in aquatic and marine environment [ 161. Is the coccoid form of 

H.pylori also an example of this? In egg passages [ 171, a classical cultivation technique, no 

coccoid bacteria were growing. 

Lately, however, Andersen et al [ 11 found by supplementing base broth medium with 2 % 

newborn calf serum, Mg2+, Cu”, Fez’, Zn”, Mn2’ and lysed human erythrocytes that cocci 

reverted to spiral forms at Ph 8.0-8. 5 with recovery of urease activity. 

In order to find out whether and when certain coccoid H.pylori are viable a vast amount of 

work with many techniques has been used. 

In order to induce coccoid forms of H. pylon in vitro several methods were scrutinized. 

36 



Different chemicals as bismuth, as well as bile acids, antibiotics, temperature, nutritients, 

oxygen tension, starvation and aging are some examples [3, 10, 15, 16,21,25]. 

Some antibiotics seem t o  create greater changes to the cell wall than bismuth salts and bile 

acids [25]. In vivo changes and transformation mechanisms are less well known. Coccoid 

H.pylori bacteria seem to be relatively more common in the duodenal mucosa than in the 

stomach in humans, [26].Several TEM studies have analyzed the transformation of spiral 

H. pylon bacteria into a coccoid form [2-5, 8, 10, 121. SEM-investigations are few [ I  5 1 . 0 ~ ~  

TEM findings of coccoid formations are in accordance with others [2-5, 8, 10, 121. Dense 

material is accumulated in the periplasmic space, the spiral bacteria bend and the outer membrane 

is separated from the inner layer. Remodeling of imer structures occurs and formation of a 

coccoid takes place. Benaissa et al[2] described remaining polar membranes and invagination 

of surface membranes. We also found flagella structures, found by several other authors [3, 101, 

but denied by others [19,25,29]. The coccoid center seems t o  contain light areas. Bode et al [3] 

demonstrated that they contain polyphosphate-rich material by electron spectroscopic imaging, 

which is used for basic energy requirements. These light areas were found by us and others [2,4, 

121. Membranes were found to be intact in most coccoids but also degenerative forms were found 

[101. 
The formation of membrane vesicles is a striking finding. Later in the coccoid development these 

vesicles are squeezed off. This might be a remodeling with reduction of surface for better survival, 

morphologic expression of detoxification mechanism or simply a degenerative phenomenon [21]. 

As the membrane vesicles seem to be a phenomenon characteristic of all coccoid formation we 

believe that this is a fbnctional rather than a degenerative event. 

The size of the coccoid seems to vary depending on the method used to produce them. Bode 

et a1 [3] found a mini-form of 0.3 pm when using antibiotics. Other methods yielded bacteria of 

the size 0.8-1.5 pm as judged by published photos [2,3, 101. Our coccoid bacteria showed the 

diameter of 0.9-1.4 pm as measured by a special device installed inside the electron microscope 

which enabled us to measure the diameter already at the screen. 

SEM showed clusters of coccoid bacteria with slight uneven surfaces. No spiral forms were 

detected after 5 days cultivation. Clustering seems to be a feature in the human stomach [I21 

as well as in vitro [ 151. 

Marked differences between the spiral and coccoid form were discovered [8, 161 by chemical 

and physiological measurements as well as immuno-, DNA, RNA and ATP-analyses. 

3 1  



A substantial amount of similarities were, however, discovered with intact or similar properties 

in both forms of H. pylori. Thus, DNA and RNA were sometimes preserved , especially in 
younger coccoid cultures [16, 18,22, 301 with incorporation of BrdU [31] as a sign of protein 

synthesis [31], a finding however denied by others [21]. Some authors have reported decreased 

RNA /DNA content with non-random fragmentation but rarely nil-values [ 16,21,23]. Levels 

of increased ATP have also been found after addition of fresh medium [30]. Specific cellular 

antigens sometimes remained [28] but a substantial modification in the cell wall of peptido- 

glycans occurred [14]. Immunoblot and RAPD-fingerprinting report loss of 160 kD, 120 KD 

(vac A) and a 35 k D cell surface protein while most other protein bands were weakened or 

preserved at normal levels [25, 30, 311. 

Adherence of H. pylori coccoid to gastric epithelia and glycocalyx was the same for cocccoid 

and spiral forms as was also staining for 8 types of lectins [27]. Cole at al. [13] reported poor 

adherence to gastric epithelial cells with little or no interleukin-8 secretion. Adhesion with 

pedestal formation was seen between H. pylori coccoid and epithelial cells [ 19, 3 11 as well as 

in a Balb/cA mice H.pylori infection model [32]. Plasminogen and lactoferrin bind to the 

coccoid form of Helicobacter, and also to extracellular matrix protein considered to be an 

important mechanism for tissue adhesion [20]. 

Moran [24] suggests that we can not exclude that two different types of H. pylori coccoid 

forms exist; one that is degenerative and one viable but not (at that time) cultivable and that 

the viable form later on can develop to a degenerative form. The possibility that a minority of 

bacteria, not detected by methods commonly used, can stay in coccoid or resting state can not 

be ruled out [ 161. This view might explain the contradictory findings present today regarding 

viability of the coccoid form of H.pylori. The advantage of TEM is that single cells are 

analyzed compared to most other methods and that l l l y  intact as well as degenerated bacteria 

can be recorded. 

We strongly believe that some of the coccoids are viable due to the fact that subcellar 

structures and intact membranes were found in our TEM and SEM studies. We were also able 

to see coccoid bacteria attached to gastric epithelia in Balb/cA mice with subsequent 

development of gastritis after 16 weeks [32] .Cellini et al. [7] observed the same in Balb/c mice. 

This l r t h e r  strengthens our view that the coccoid form of H. pylon is viable. 

38 



ACKNOWLEDGEMENT 

This study was supported by grants from the Swedish Medical Research Council (16 x 04723), 
Royal Physiographic Society, Lund and founds from The Sahlgrenska University Hospital, 
Goteborg, Sweden. The skilled photographic help of Michael Pettersson is greatly appreciated 

REFERENCES 

1. Andersen, A. P., Elliott, D. A,, Lawson, M., Barland, P., Hutcher, V.B.& Puszkin, E.G.: 
Growth and morphological transformation of helicobacter pylori in broth media. J Clin 
Microbiol35: 2918-2922,1997. 

2. Benaissa, M.,Babin, P., Quellard, N., Pezennect, L., Cenatiempo, Y.& Fauchere, J. L.: 
Chmges in he!icobacter pylori dtrastmcture end antigens during conversion fiom the 
bacillary to the coccoid form. Infect immun 64:2331-2335,1996 

3. Bode, G., Mauch, F.& .Malferthheimer, P. :The coccoid forms of helicobacter pylori. 
Criteria for their viability. Epidemiol infect 1 1 1 :483-490 , 1993. 

4. Caselli, M.,Aleotti, A,, Boldrini, P., Ruina, M.& Alvisi, L.: Ultrastructural patterns of 
helicobacter pylon: Gut 34:1507-1509, 1993 . 

5. Caternich,C.E.&Mankin,K.M.: Charactrerization of the morphologic conversion of helico- 
bacter pylori from bacillary to coccoid forms. Scand J Gastroenterol supl. 18 1 : 58-64,199 1. 

6. Cave,R.:How is helicobacter pylon transmitted?Gastroenterology 113:s 9-S14,1997. 
7. Cellini,L.,Allocati,N.&Angelucci,D.: Coccoid helicobacter pylori not culturable in vitro 

8. Cellini,L. Allocati,N.,Di Campli,E.&Dainelli,B. :Helicobacter py1ori:A fickle 

9. Cellini,L.:Coccoid forms of helicobacter pylori.J.Infect.Dis. 175: 1288-1289,1996. 
10. Cellini, L., Robuffo, I., Di Campli, E., Di Bartolomeo, S., Taraborelli, T.& Dainelli, B.: 

Recovery of helicobacter pylori ATCC 43504 from a viable but not culturable state; re- 
growth or resuscitation? APMIS 106 :571-579,1998 . 

Marzio, L.: Evidence for an oral-faecal transmission of helicobacter pylori infection in an 
experimental murine model. APMIS 107 :477-484, 1999 . 

12.Chan, W. Y., Hui, P.K., Leung, K.M., Chow, J., Kwok, F.& Ng, C. S.: Coccoid forms of 
helicobacter pylon in the human stomach. Am J Clin Pathol 102 :503-507,1994 Cellini, L., 
Robuffo, I., Di Campli, E., Di Bartolomeo, S., Taraborelli, T.& Dainelli, B.: Recovery 

13 .Cole, S. P., Cirillio, D., Kagnoff, M.F., Guineg, D.G.& Eckmann, L.: Coccoid and spiral 
helicobacter pylori differ in their abilities to adhere to gastric epithelial cells and induce 
interleukin- 8 secretion. Infect immun 65:843-846,1997 . 

14.Costa, K., Bacher, G., Allmaier, G., Dominguez-Bello, M.G., Engstrand, L., Falk, P., de 
Pedro, M. A. & Garcia-del Portillo, F.: The morphological transition of helicobacter pylon 
cells from spiral to coccoid is precided by a substantial modification of the cell wall. J 

Bacteriol 181 :3710-3715, 1999. 

Bartolomeo, S. & Cellini, L.: The effect of oxygen on the growth and cell morphology of 
helicobacter pylori: FEMS Microbiol lett 168:9- 15 , 1998 . 

16.Enroth, H., Wreiber, K., Rigo, R., Risberg, D., Uribe, A. & Engstrand, L.: In vitro aging 
of helicobacter pylori: Changes in morphology, intracellular composition and surface 
properties. Helicobacter 4: 7- 16,1999. 

reverts in mice.Microbio1 Immunol38:843-850,1994. 

germ.Microbiol.Immuno1.3 8:25-30,1999, 

11 .Cellini, L., Dainelli, B., Angelucci, D., Grossi, L., Di Bartolomeo, S., Di Campli, E.& 

lS.Donelli, G., Matarrese,P., Fiorentini, C., Dainelli, B., Taraborelli, T., Di Campli, E., Di 

39 



17.Enroth, H. & Engstrand, L.: Egg passage of rodshaped and coccoid forms of helicobacter 
pylori: Preliminary studies. Helicobacter 1: 183-186,1996. 

18.Hua, J. & Ho, B.: Is the coccoid form, of helicobacter pylon viable? Microbios 87 : 103 -1 12, 
1996. 

19.Janas, B., Czkwianianc, E., Bak-Romaniszyn, L., Bartel, H., Tosik, D. & Planeta-Malecka, I.: 
Electron microscopic study of association between coccoid forms of helicobacter pylori and 
gastric epithelial cells. Am J Gastroenterol. 90 : 1829-1833 ,1995. 

20.Khin, M.M., Ringner, M., Aleljung, P., Wadstrom, T., Ho, B.: Binding of human plasmino- 
gen and lactofemn by helicobacter pylori coccoid forms. J Med. Microbiol:45:433-439, 1996. 

21.Kusters, J.G., Gemts, M.M., van Strijp, J.A. & Vanen Broucke-Grauls, C.M.: Coccoid forms 
of helicobacter pylon are the morphologic manifestation of cell death. Infect immun. 65:3672- 
3679 ; 1997. 

22.Mizoguchi, H., Fujioka, T., Kishi, K., Nishizono, A,, Kodanma, R. & Nasu, M.: Diaversity 
in protein synthesis and viability of helicobacter pyiori cowoid forins in response to various 
stimuli. Infect immun. 66:5555-5560,1998, 

23.Monskins, H.J., Tiveljung, A. & Jonasson, J.: Non-random fragmentation of ribozomal RNA 
in helicobacter pylori during conversion to the coccoid form. E M S  Immunol. Med. Microbiol. 
22:217-224,1998. 

24.Moran, A. P.: Coccoid forms of helicobacter pylori. Helicobacter 2: 109-1 10,1997. 
25.Nilius, M., Strohle, A., Bode, G. & Malfertheimer, P.: Coccoid like forms (CLF) of helico- 

bacter pylori. Enzyme activity and antigenicity. Zentralbl.Bakterio1. 280 :259-272,1993. 
26.Noach, L. A,, Rolf, T. M. & Tytgat, G.N.: Electron microscopic study of association between 

helicobacter pylon and gastric and duodenal mucosa. J. C h  pathol47:699-704,1994. 
27.0gata, M., Araki, K. & Ogata, T.: An electron microscopic study of helicobacter pylori in the 

surface mucosa gel layer. Histol Histopathol 13;347-358 ,1998. 
28.Roi, I.H., Son. S.H. 0. H. H.T., Choi, J. Shin, J.H., Lee, J.H. & Hah, Y.C.: Changes in the 

evolution of antigenic profiles and morphology during coccoid conversion of helicobacter pylon. 
Korean J Intern. Med. 14:9-14,1999. 

induces cytoskeletal rearrangements and tyrosine phophorylation of host cell proteins.Proc 
Natl Acad Sci 93: 1259-1264,1996. 

3O.Sorberg, M., Nilsson, M., Hanberger, H. & Nilsson, L.E.: Morphologi conversion of helico- 
bacter pylon from bacillary to coccoid form. Eur. J. Clin. Microbiol. Infect Dis. 15:216-219, 
1996. 

31.Vijayakumaris, S., Khin, M.M., Jiang, B. & Ho, B.: The pathogenic role of the coccoid form 
of helicobacter pylori. Cytobios 82:25 1-260 ,1995. 

32.Wang, X., Stureghd, E., Rupar, R., Nilsson, H.O., Aleljung, P.A., Carlen, B., Willen,R.& 
Wadstrom T: Infection of BALB IcA mice by spiral and coccoid forms of helicobacter 
pylon. J. Med. Microbiol46:657-663 ,1997. 

29.Segal, E.D., Falkow, S.& Tompkin, L. S.: Helicobacter pylon attachment to gastric cells 

Offprint request to: Roger Willen, MD , Ph.D. 
Department of pathology 
Sahlgrenska university hospital 
S-413 45 , Gothenburg, Sweden. 

40