British Journal of Dermatology 2001; 145: 114±116.

Identification of Mycobacterium marinum in sea-urchingranulomas

C.DE LA TORRE, A.VEGA,* A.CARRACEDO² AND J.TORIBIO³

Service of Dermatology, Complejo Hospitalario de Pontevedra, C/Loureiro Crespo s/n, 36001 Pontevedra, Spain

*Unidad de Medicina Molecular, SERGAS, Santiago, Spain

²Instituto de Medicina Legal and ³Departamento de DermatologiÂa, Universidad de Santiago, Santiago, Spain

Accepted for publication 23 February 2001

Summary Background Sea-urchin granuloma is a chronic granulomatous reaction arising after injury withsea-urchin spines. Classified as an allergic foreign-body type of granuloma, it is believed to be a

delayed-type reaction to an as yet unidentified antigen. In a clinicopathological study, 50 biopsyspecimens from 35 patients diagnosed as having sea-urchin granuloma caused by Paracentrotus

lividus, we found different inflammatory patterns that in some cases suggested a mycobacterial

infection.Objectives To investigate and identify mycobacterial DNA in formalin-fixed and paraffin-embedded

skin biopsy specimens diagnosed as sea-urchin granulomas.

Methods A search combining polymerase chain reaction amplification using Mycobacteriumgenus-specific primers, and subsequent restriction enzyme analysis enabling identification to the

species level, was performed in 41 samples.

Results Amplification of a 924-bp DNA fragment encoding mycobacterial 16S rRNA gene waspositive in eight biopsy specimens from seven patients (21%). M. marinum-specific restriction

patterns were identified in three samples.

Conclusions Although further controlled studies are necessary, from these data it would appearthat myobacteria may play a pathogenic role in some cases of sea-urchin granuloma.

Key words: echinoderm dermatitis, marine injuries, mycobacteria, Mycobacterium marinum,Paracentrotus lividus, polymerase chain reaction, sea-urchin granuloma

Sea-urchins can produce lesions either with the spines

or the pedicellariae. Two types of sea-urchin reactionshave been described: immediate and delayed reactions.

Delayed reactions may be nodular or diffuse and are

designated `sea-urchin granulomas'.1±5 The sea-urchinParacentrotus lividus, common in Mediterranean and

Atlantic coastal waters, is the main culprit in thereported cases of sea-urchin granuloma.6±8 To date,

investigations have failed to demonstrate the causative

agents able to induce this granulomatous reaction.

Materials and methods

A histopathological study was carried out using 50

biopsy specimens from 35 patients. Using archival

material from this study, polymerase chain reaction

(PCR) was performed in 41 formalin-fixed and paraffin-embedded skin biopsy specimens. Tissue was not

available in nine instances, but PCR was performed at

least once in every patient. We used PCR withMycobacterium genus-specific primers encoding small

subunit rRNA sequences, and digestion of the amplifiedfragment with restriction enzymes (BanI and ApaI) that

yielded a specific pattern enabling identification of

M. marinum.9,10

Specimens were cut to 10 mm and deparaffinized

with 1 mL xylene for 1±2 h and then centrifuged for

15 min. Measures were installed to avoid cross-contamination, including the use of dedicated labora-

tory areas. Only one item from a case was analysed in

the laboratory at a time, and reagent blanks andnegative controls were used through the process. The

samples were washed in an ethanol series (100%, 80%,

114 q 2001 British Association of Dermatologists

Correspondence: Carlos De la Torre, Plaza AmeÂrica 3±58 D, 36211

Vigo, Spain. E-mail: ctorre@arrakis.es

M. MARINUM IN SEA-URCHIN GRANULOMAS 115

q 2001 British Association of Dermatologists, British Journal of Dermatology, 145, 114±116

70%). The pellet was dried and resuspended in 100 mL

digestion buffer (200 mg mL21 proteinase K,50 mmol L21 Tris hydrochloride pH 8´5, 1 mmol L21

ethylenediamine tetraacetic acid, 0´5% Tween 20) at

56 8C for 2 h and then the proteinase K inactivated byboiling for 10 min11 Two oligonucleotide primers were

used to amplify a fragment of 924 bp from the 16SrRNA gene of Mycobacterium9 (P1, GCGAACGGGT-

GAGTAACACG; P2, TGCACACAGGCCACAAGGGA).

DNA amplification was carried out in a 25-mL volumewith 5±25 ng template DNA, 10 mmol L21 Tris

hydrochloride pH 8´3, 50 mmol L21 KCl, 1´5 mmol L21

MgCl2, 200 mmol L21 of each deoxyribonucleosidetriphosphate, 10 pmol of each primer and 1´25 U of

Taq polymerase (Life Technologies, Rockville, MD,

U.S.A.). The reactions were carried out in a thermalcycler (Progene, Techne, Princeton, NJ, U.S.A.), and

consisted of 30 cycles with three steps of denaturation

(95 8C, 30 s), annealing (55 8C, 30 s) and extension

(72 8C, 45 s). The amplified DNA products weredigested by restriction enzymes ApaI and BanI (Life

Technologies), and the digested samples were analysed

by electrophoresis on polyacrylamide gels (PhastGels,APB, Uppsala, Sweden) and stained with silver nitrate.

All the positive samples were analysed at least three

times each, with consistent results.

Results

The cohort consisted of 35 patients (31 male and four

female) diagnosed as having sea-urchin granulomas

(Fig. 1). The patients had a median age of 35 years(range 14±60), and a median latency time from the

time of injury of 7´5 months. Half of these patients

were involved in fishing activities, and 50% of thesewere divers dedicated to commercial harvesting of

sea-urchins.

Fifty biopsy specimens were available for histologicalstudy. Several inflammatory patterns were noted. In

70% there was a granulomatous pattern, mainly of

sarcoidal (n � 13; 26%) or foreign-body (n � 10; 20%)type. Other granulomatous patterns identified were

tuberculoid (n � 2), necrobiotic (n � 6) and suppura-tive granulomas (n � 4). In 30% the predominant

pattern was non-granulomatous with features of

chronic and suppurative dermatitis. Acid-fast bacilliwere not identified using Ziehl±Neelsen stains. Skin

biopsy specimens were cultured in only 11 instances,

none of which gave positive isolates of mycobacteria.A PCR-amplified DNA fragment of 924 bp, encoding

mycobacterial 16S rRNA, was obtained in eight

biopsies from seven patients. Digestion yielded threefragments (677 bp, 132 bp, 115 bp) specific for

M. marinum9 in three of these samples (Fig. 2). The

remaining patterns, with two fragments (792 bp and132 bp) were not specific for M. marinum, suggesting

the possibility of a mutation, subspecies or other non-

identified mycobacterial species. Among the positiveresults, four cases showed predominantly granuloma-

tous inflammation. Statistical analysis (x2-tests)

showed no correlation between positivity and patternof inflammation.

Discussion

The main constituent of sea-urchin spines is calcium

carbonate, which is considered immunologically inert.It has, therefore, been postulated that a proteinaceous

compound generated by the spine epithelium, a

Figure 1. Sea-urchin granuloma with papular and inflammatory

lesions involving the dorsum of the fingers.

Figure 2. Polymerase chain reaction from cutaneous samples of sea-

urchin granuloma. Left: amplified 924-bp DNA fragment encoding

Mycobacterium 16S rRNA. Right: restriction enzyme analysis withApaI. Lanes 1, 2 and 6 show a pattern specific for M. marinum.

116 C.DE LA TORRE et al.

q 2001 British Association of Dermatologists, British Journal of Dermatology, 145, 114±116

substance that can enter the wound, either produced

by the pedicellariae or adherent to the spine (sand,slime, algae, etc.) could be the agent responsible for

sensitization. Usually no microorganisms, sarcoidal

diathesis or metals likely to produce granulomas havebeen detected; the spine silica content is about 0´55%,

but no polarized material has been identified in

biopsies.5,6,12 The first description13 suggested papulo-necrotic tuberculide, and although a mycobacterial

infection was sought,7 acid-fast bacilli were detected

in only one case14 in which the authors proposedthat echinoderm granuloma could be a new myco-

bacterial infection. The reaction was later defined as

sarcoidal.8 We cannot find any reports in thedermatological or rheumatological literature docu-

menting a systematic search for mycobacteria in

cases of this condition.As M. marinum is not a commensal, and contamina-

tion is unlikely, its identification favours a pathogenic

role in some cases of sea-urchin granuloma. As thiswas not identified in all cases, it is likely that several

different aetiopathogenic mechanisms may be involvedin the granulomatous lesions provoked by punctures

with sea-urchin spines. The spines can occasionally

harbour M. marinum, and Mycobacterium is probablyone of the implicated agents. Although this is a

tentative and preliminary conclusion we think that it

may have pathogenic and therapeutic implications tobe investigated further.

In addressing future studies on this topic, it should

be remembered that the diagnosis of cutaneousM. marinum infection is frequently presumptive, as

detection by conventional methods is difficult,15 and

also that several Mycobacterium species that arepathogenic to humans can infect aquatic species.9

References

1 Rocha G, Fraga S. Sea urchin granuloma of the skin. Arch

Dermatol 1962; 85: 146±8.

2 Warin AP. Sea-urchin granuloma. Clin Exp Dermatol 1977; 2:

405±7.

3 Fisher AA, Orris WL. Aquatic contact dermatitis. Cutis 1973; 12:

687±95.

4 Haneke E, Kolsch I. Seeigelgranulome. Hautarzt 1980; 31:

159±60.

5 Baden HP. Injuries from sea urchins. Clin Dermatol 1987; 5:112±17.

6 Anonymous. On the rocks. (Editorial) Lancet 1974; i: 1091±2.

7 Kinmont PDC. Sea-urchin sarcoidal granuloma. Br J Dermatol

1965; 77: 335±43.

8 Cooper P, Wakefield MC. A sarcoid reaction to injury by sea

urchin spines. J Pathol 1974; 112: 33±6.

9 Talaat AM, Reimschuessel R, Trucksis M. Identification of

mycobacteria infecting fish to the species level using polymerase

chain reaction and restriction enzyme analysis. Vet Microbiol1997; 58: 229±37.

10 Tùnjun T, Welty DB, Jantzen E, Small PL. Differentiation of

Mycobacterium ulcerans, M. marinum, and M. haemophilum:mapping of their relationships to M. tuberculosis by fatty acid

profile analysis, DNA-DNA hybridization, and 16S rRNA gene

sequences analysis. J Clin Microbiol 1998; 36: 918±25.

11 Sobrido MJ, Barros F, Lema M et al. Assessing microsatellite

instability with semiautomated fluorescent technology. Applica-

tion to the analysis of primary brain tumours. Electrophoresis2000; 21: 1471±7.

12 Asada M, Komura J, Hosokawa H et al. A case of delayedhypersensitivity reaction following a sea urchin sting. Dermato-

logica 1990; 180: 99±101.

13 Gate J, Cuilleret P, Chanial G, Bouquin H. LeÂsions papulo-neÂcro-tiques aÁ reÂaction histologique tuberculoiÈde dues aÁ l'inclusion

d'eÂpines d'oursins. Bull Soc Fr Dermatol Syphiligr 1936; 43: 937±8.

14 Moynahan EJ, Montgomery PR. Echinoderm granuloma: a skin

lesion resulting from injury by the spines of sea-urchins

inhabitating temperate waters? A new mycobacterial infection.

Br J Clin Pract 1968; 22: 265±9.

15 Ang P, Rattana-Apiromyakij N, Goh C-L. Restrospective study of

Mycobacterium marinum skin infections. Int J Dermatol 2000; 39:343±7.