Pneumonia caused by a previously undescribed bacterium

(1)

0095-1137/82/090536-06$02.00/0

Pneumonia Caused

by a

Previously Undescribed Bacterium

ROYL. HOPFER,l* KAREN MILLS,' VICTORFAINSTEIN,2HARALD E. FISCHER,' AND MARIOP. LUNA3

Departmentsof LaboratoryMedicine,1 DevelopmentalTherapeutics,- andPathology,3 University of Texas M.D.Anderson Hospitaland Tumor Institute, Houston, Texas 77030

Received 12February1982/Accepted21 May 1982

Anew and as yet unidentified bacterium was isolated from the lung tissue of a

cancerpatient with bilateralpneumonia. Clinically, thepneumoniawasconsistent

withlegionellosis; the organism cultured from the lung grew only on the

charcoal-yeast extract agar routinely used for Legionella isolation. Subsequent testing,

however, showed the organism to be quite distinct from the known Legionella

species in itsbiochemical, antigenic, and growth characteristics.

Charcoal-yeast extract agar(CYEA) has been

widely used as a primary isolation medium for

Legionella pneumophila and related organisms

since its description byFeeley et al. (4).

Gram-negative bacilli cultured on CYEA that fail to

grow on other commonly used bacteriological

media havebeengenerallyreferred to as

Legion-ella-like organisms. Once such organisms are

fullycharacterized by biochemical testing, DNA

hybridization, morphology, and antigenic

relat-edness, they aregiven new scientific names or are assigned to different serogroups within the same species.

Todate, six distinct serogroupsofL.

pneuimo-philahavebeenidentified,as havefive

addition-al Legionella species (2, 8, 13, 14). Two new

genera (Fluoribacter and Tatlockia) have been

proposed (6), which would replace L. bozemnanii

and L. micdadei as F.bozemanae and T.

micda-dei, respectively.

Weisolated agram-negativepleiomorphic rod

fromlung tissue obtainedatautopsyof a cancer

patientwhodied ofabilateralpneumonia.

Clini-cally, the pneumonia resembled legionellosis,

and the cultured microorganism grew only on

the CYEA mediumcommonlyusedfor isolation

ofLegionella species. The isolation of this

mi-croorganism was fortuitous, since it grew only

within oraround colonies of Candida

parapsilo-sis. Because the microorganism produces a

pink, nondiffusable pigment, it will be referred

to as Pink-MDA.

(This paper was presented in part at the 81st

Annual Meeting of the American Society for

Microbiology, 6 to11 March 1981, Dallas, Tex.)

CASE REPORT

A 54-year-old woman was diagnosed as having

squamous cell carcinoma of the esophagus in May

1979. She had ahistoryof heavycigarette smokingfor 40 years, and, on physical examination, had chronic

bronchitisand severe emphysema. In June 1979, she

received radiation therapy to herchest. However, 2 months later she developed metastases to the scalp andbrain,whichweretreatedby radiation. In October 1979, she suddenly developed fever and cough, and her condition rapidly deteriorated. On admission to thehospital, she was acutely ill with severehypoxemia anda left lower lobe infiltrate. She was immediately

intubated andreferredtotheintensive care unit.Two days later she developed bilateral basal pneumonia

withpleural effusionsdespiteintensiveantibiotic treat-mentwith intravenous penicillin and tobramycin. On

day 4, erythromycin was added to the therapeutic regimen, butthepatient diedthat sameday because of severerespiratory insufficiency.

Onpostmortemexaminationbothlungs hada granu-lar appearance. The tissue was firm and showed a

reticular pattern.The right lungweighed 1,010 g and the left, 1,510 g. Histological examination disclosed

expansion ofthe alveolar spaces filled with necrotic

debris and many polymorphonuclear leukocytes and extravasation of erythrocytes into the alveolar and

bronchiolarspaces. Theentire leftlung was involved. The alveolar septa showed congested capillaries and

fibrosis. Gram stain of the tissue revealed

gram-negative bacilli. Special stainsand cultures for

acid-fastbacilli andfungiwerenegative.

MATERIALS AND METHODS

Growth studies.Pink-MDA grewinitially only with

colonies of C. parapsilosis. Themicroorganism grew

onlyonCYEA, althoughthe lungspecimen had been inoculated onto a variety of other commonly used

bacteriological, fungal, and mycobacteriological

me-dia.Therefore,allsubsequentgrowthstudies included a symbiotic culture with C. parapsilosis. The yeast was added directly to liquid medium and cross-streaked orstreaked in closeproximitytoPink-MDA. As these studies progressed, other microorganisms (seebelow)werefound that couldreplace C.

parapsi-losis as the symbiotic organism. Attempts were made togrow Pink-MDA insymbioticculture on most of the commonly available microbiological media (primary isolation, enrichment,and selectivemedia)and on the

536

on February 7, 2020 by guest

http://jcm.asm.org/

(2)

CAUSED BY A NEW BACTERIUM 537

highlyenriched media used for isolation of fastidious

microorganisms.These cultureswereincubated under aerobic and anaerobic conditions, in 5 to 10% CO,. andin candle jarsatboth 25 and 35°C. However, Pink-MDA had to be maintained in symbiotic culture on

CYEA.

Subsequent nutritional requirement studies were

carriedout withCYEAoryeastextractbroth(18)or

modifications ofthese media. Avariety ofvitamins,

cofactors, amino acids,carbohydrates,minerals, salts, etc.,wereeither addedtoordeletedfromtheCYEA basal medium. Variouscomponentsof CYEA medium

(charcoal, yeast extract, L-cysteine, ferric pyrophos-phate) were omitted or filter sterilized rather than being autoclaved. In additiontothecommon biochem-ical tests(such ascatalase, oxidase, etc.) that donot

require growth ofthe organism, Pink-MDAwasalso

tested in the API ZYM (AnalytabProducts,Plainview, N.Y.). Carbohydrate utilization studies were

per-formed with folic acid-supplemented yeast extract

broth.

Serological studies. (i) Fluorescent microscopy. The method of McKinney et al. (14), as modified by Lewallenetal.(9), was usedtoproduce antiserumto

Pink-MDA in rabbits. Indirect fluorescent antibody

(IFA) testing was performed as follows. Smears of

Formalin-killed test organisms were air dried, heat

fixed,allowedtocool, coveredwith20,ulofantiserum

toPink-MDA, and allowed toincubate for 30 minat room temperature in a moist chamber. After being

washed inphosphate-bufferedsaline (pH7.6, 0.01 M), 20 ,u1 ofcommercially available fluorescein isothio-cyanate-labeled goat anti-rabbit immunoglobulin G (IgG) (Cordis Laboratories, Miami, Fla.) was added

andallowedtoincubate for 30min. After anotherwash inphosphate-buffered saline, the slideswereexamined

under a fluorescent microscope. The Formalin-fixed

L. pneumophila (serogroups 1-6), L. micdadei, L. gormanii, L.dumoffii, andL. bozemanii (Centers for Disease Control[CDC], Atlanta,Ga.)werealso tested by the IFA methodwith the Pink-MDA antiserum in the initial coating step. In addition, other bacteria

(Table 1) were tested by the same IFA method.

Similarly, theCDC antisera (fluorescein isothiocyan-ate conjugated) against the same Legionella antigens were investigated by a direct fluorescent antibody

method against Pink-MDA. Finally, lung tissue im-pression smearsand sections from the same patient were stained by the IFA method with Pink-MDA reagentsandby the directfluorescent antibody meth-od with the Legionella reagents available from the CDC.

(ii) ELISA. All of the Legionella antigens, Pink-MDA, andavariety of other bacterial antigens (Table 1)were used to coatplastic cuvettes (Gilford

Instru-ments, Oberlin, Ohio)fortheenzyme-linked

immuno-absorbent assay (ELISA). Antiserum to Pink-MDA

wasadded andincubated for60min with commercially available alkalinephosphatase-labeledgoatanti-rabbit IgG (MilesLaboratories, Inc.,Elkhart, Ind.).Enzyme activity wasmeasured withaGilfordInstrument EIA

PR50.

DNA base composition. Cells were harvested from

the surface of folicacid-supplemented CYEA plates, washed, and lysed as previously described (10) to prepareaDNA-containingsolution. Thebuoyant

den-TABLE 1. Bacteriaandbacterialantigenstested for

cross-reactivitywith Pink-MDA antiserum

Pink-MDA Homologous Bacterium/antigen' antiserum antiserum

IFA ELISA IFA ELISA

Pink-MDA 4+ + NAb NA

L.pneumophila - - 4+ +

(serogroups1-6)c

L. micdadei - - 4+ +

L. gormanii - - 4+ +

L. dumoffii - - 4+ +

L.bozemanii - - 4+ +

E. coli - - NDd ND

Serratia marcescens - - ND ND

Klebsiella pneumoniae - - ND ND

Pseudomonas - - ND ND

aeruginosa

Corynebacterium spp. - - ND ND

Mycobacterium kansasii - - ND ND

aWhole cells andculture filtratecoated onto plastic cuvettesforELISA.

bNA, Not applicable.

C All six serogroups were tested and reacted only

withtheirrespective homologous antisera.

dND, Not done; homologous antiserum not

avail-able.

sity of the DNA in neutral CsCl was determined in

duplicate(11),with DNA ofbacteriophage2C

(Bacil-lus subtilis host) used as an internal reference at a

density of 1.742 g cm-3, relative to the primary referenceofEscherichia coliDNA (P = 1.7100; 51.0 mol% G+C). Base composition was calculated from the equation ofSchildkraut et al.(19).

RESULTS

Growth studies.Pink-MDAgrewneither alone

norsymbioticallyonanygrowth mediumexcept

CYEA. Many bacteria, including a variety of gram-positive and gram-negative organisms and yeasts, could replace C. parapsilosis as the

symbiote (Fig. 1). Symbiote culture with either

Staphylococcus aureusorE. coli allowed

colo-nies of Pink-MDA to be detected earlier than

withmostotherorganisms. Inaddition, both the

colony size and pigmentation of Pink-MDA

in-creased when grown with these bacteria rather

than withyeasts.Furthermore, when E. coliwas

allowedtogrow on CYEA for 48 h and theagar

block containing the colony ofE. coliwasthen

removed, the surrounding medium supported

the growth of Pink-MDA. Pink-MDA grew as

well under these conditions as it did in the

presenceof E. coli. These data indicate that the

growth factor(s) provided by the symbiote was

diffusable and stableat the incubation

tempera-ture. Pink-MDA failed to grow on CYEA

pre-paredat final pH values ranging from 5.5 to8.5

(testedat0.5increments). Subsequently,a

on February 7, 2020 by guest

http://jcm.asm.org/

(3)

. i_

; X

I.,

FIG. 1. Growth of Pink-MDA ("T" streak) when inoculated in close proximity to S. aureus (circular

cluster ofcolonies) onCYEA. Note luxuriantgrowth ofPink-MDAclosetoS. aurelusandincreasingly less growth with distance.

etyofliquid mediaconditioned by growth ofE.

coli or S. aureus for 24, 48, or 72 h were

examined as a replacement for the symbiote.

None of theseconditioned media could support

thegrowthof Pink-MDAonCYEAoranyother

solid medium.

Pink-MDA would notgrow onCYEA alone if

the agar, yeast extract, and charcoal had been autoclaved. Filter sterilization of the yeast

ex-tract allowed some growth of Pink-MDA

with-outthe symbioteregardlessof whethercharcoal

waspresent(Table 2).However, subculturing of the colonies to another filter-sterilized CYEA

plate resulted in no detectable growth of

Pink-MDA. Similargrowthresultswereobtained with

filter-sterilized yeast extract broth. Pink-MDA tendedtobepleiomorphicand produced

round-edspheroplast-likecells inyeastextractbrothor

other liquid media (Fig. 2). The addition of autoclaved yeast extract, but not autoclaved

agarblocks, totheyeast extract broth inhibited

thegrowthof Pink-MDA. The individual

compo-nentsof yeast extract (Table 2) were then tested

for their ability to support the growth of

Pink-MDA on CYEA. Although a Casamino Acid

solution allowed some growth of

Pink-MDA,

folic acidappearedtobe the single best

growth-promoting factor forPink-MDA on CYEA or in

yeast extract broth. Rounded cells were also

detected in folic acid-supplemented yeast

ex-tract broth. In the presence of folic acid,

Pink-MDA grewas well as or better than it did with

any symbiote and showed intense pink

pigmen-tation. Interestingly, autoclaving the folic acid

before adding it to theCYEA did not destroyits

capacity to support the growth of Pink-MDA.

Incorporating1,10, or 100 pLg of folic acid per ml

into CYEA allowed excellent growth of

Pink-MDA; however, all concentrations of folic acid seemed to inhibit the growth of L. pneiumophila

(serogroups 1, 3, and 6).

Thebiochemicalcharacteristics of Pink-MDA

compared with those of the Legionella species

areshown in Table 3. Pink-MDA is anonmotile,

catalase- and

oxidase-negative,

gram-negative

rod that produces a pink pigment. Electron

microscopy (Fig. 3) of Pink-MDA revealed an inner and outer cell wall structure typical of

gram-negative organisms. The guanine plus

cy-tosine (G+C) content of Pink-MDA was 62 mol%.

Serological studies. The antiserum raised

against Pink-MDA reactedonly with Pink-MDA

and notwith anyof the otherbacteria tested by

the IFA method(Table 1). Similarly,only

autol-ogous antisera gave apositive reaction with the

Legionella antigens in the direct fluorescent

ai,tibody test, whereas none of the Legionella

antisera stained Pink-MDA by the same

proce-dure. Increasingthesensitivity of thetestbyan

ELISAmethod didnot changethe results, i.e.,

Pink-MDAappeared tobe antigenicallydistinct

from all ofthe other bacteria tested(Table 1).

Fluorescent bacteria were only seen in thin

sections oflungtissue fromourpatientwhen the

tissue wasstained with antiserumraisedagainst

Pink-MDA. None of the CDC fluorescent

anti-sera gave positive microscopic findings in the

tissue sections.

DISCUSSION

Pink-MDAwasisolated fromacancerpatient

with bilateral basalpneumonia. Organisms were

seen by Gram stain, fluorescence

microscopy,

and the Steinert and Steinert silver stain. Tissue

pathologywasconsistent withlegionellosis.The

etiological agent appears to be a previously

undescribedorganism.

We were very fortunate in isolating Pink-MDA. The organism grew only on a medium

designedforisolation of theLegionella species.

CYEA alone does not support the growth of

J. CLIN. MICROBIOL.

on February 7, 2020 by guest

http://jcm.asm.org/

(4)

TABLE 2. Growth of Pink-MDA on a variety of media

Growth"

Medium Treatment orsupplement" Without With

symbiote symbiote' Routine media (20 commonly used None

media)

CYEA None"d 4+

YEA No

charcoald-CYEA-FC Noferricpyrophosphate or 4+

L-cysteine

FS-CYEA Filter sterilizedyeast 2+e 4+

extract

FS-YEA Filter sterilized, no 2+e 4+

charcoal

FS-YEB Filter sterilized yeast 2+± ND

extract broth

FS-YEB-AA FS-YEB withautoclaved 2+e ND

agar

CYEA CasaminoAcid 1+e ND

CYEA Carbohydrates (dextrose,

sucrose, maltose, ND

xylose, raffinose, arabinose, trehalose)

CYEA Peptone - ND

CYEA Biotin - ND

CYEA Folic acid 4+f ND

CYEA Pantothenic acid ND

CYEA p-Aminobenzoic acid ND

CYEA Riboflavin ND

CYEA Thiamine ND

CYEA Niacin ND

CYEA Thymidine 4+' ND

CYEA Gey balanced salt solutionl ND

aSupplements were filter sterilized.

b ,No growth; 4+, large colonies within 3days,2+, small colonies within 5days;1 +, barely visible colonies within 5 days; ND, not done.

C S. aureus inoculated in close proximity.

dOnlyferricpyrophosphate andL-cysteine hydrochloride werefilter sterilized.

eCould not begrownwhen subculturedto the same mediumunless symbiotewas supplied.

fCouldgrowwhensubcultured to same medium withoutsymbiote. g GIBCO Diagnostics, Grand Island, N.Y.

Pink-MDA; however, Pink-MDAgrewwith

col-onies of C.parapsilosis. The symbiotic growth

of Pink-MDA and C. parapsilosis could not be

demonstrated on any other growth medium.

Many bacteria and fungi were able to provide

theneeded growthfactor(s) in symbiotic culture.

Because attempts at symbiotic cultureon other

growth media were not successful, more than

onefactormay begrowthlimiting. Furthermore,

substances inhibitory tothedevelopment of the

bacterium on CYEA and other growth media

mayhave tobe considered.

This "toxic" agent is associated with

auto-claving the yeast extract, since Pink-MDA can grow on folic acid-supplemented CYEA if fil-tered sterilization is used to prepare the yeast extract. Furthermore, the charcoal, which

prob-ably binds the inhibitor, is not needed if the

yeast extractis filter sterilized. Studies are

con-tinuingtoelucidate thisphenomenon. There is a

definite growth factor requirement for

Pink-MDAthat isnotsuppliedby CYEA. Our

prelim-inary studies indicate that folic acid is required forgrowth ofthebacterium. Failuretogrow on CYEAmight be due to alimited supply offolic

acid in the medium. Thymidine couldsubstitute

for folic acid in supporting thegrowth of

Pink-MDA, indicating a possible disturbance in the

formation ofthymidine-5-phosphate. This

phe-nomenon could be explained by a lack of folic

acid,which is involvedas amethylgroupdonor

in the synthesis of thymidine-5-phosphate (7).

Thymidine-dependent strains of

Enterobac-teriaceae are known to occurafter sulfonamide

treatment(1, 10, 16, 21). In thiscase, however,

neither sulfonamide nor Amethopterin

(metho-trexate), an antifolate, were used to treat the

patient.

VOL. 16,

on February 7, 2020 by guest

http://jcm.asm.org/

(5)

540 HOPFER ET AL.

4-*.S

6-:11|r

1'0

1is%

I

_ # ,@

I

.1

pj

4

-:1

-_

ae

QF-wa %*

K.

*9~~~~~~~~~ 0

:,*

10

* l:

..$:

o

FIG. 2. Gram stain of Pink-MDA in yeast extract brothshowing pleiomorphicand rounded

spheroplast-like cells.

TABLE 3. Comparison ofPink-MDA with

Legionella species

Test Legionella Pink-MDA

species

Gram stain Gram Gram

negative negative

Flagella Positive Negative

Catalase Positive Negative

Oxidase Variable Negative

Beta-lactamase Variable Negative

Growthrequirements

Folicacid(thymidine) No Yes

Cysteine Yes No

Iron Yes No

API ZYMa

Lipase Negative Positive

Trypsin Negative Positive

Valine Positive Negative

Growth temp(°C)

22 Positive Negative

37 Positive Positive

G+C content(mol%) 38.8-44.3 62

aOnly the three enzymes that weredifferent from thoseof theLegionella speciesarelisted.

'P~~~~~~~~~~~~~~$

>t , > e s .o~~~~~t

.e~~~~~~%

FIG. 3. Electron micrograph ofPink-MDA

show-ing typical gram-negative cell wall. Bar, 1

p.m.

The

inability

of the

Legionella

species

togrow

on

commonly

used

bacteriological

media is well

known, and several media have been described

for the culture of these related

organisms

(3, 5,

6, 17, 20, 22).

Microbiologists

have

long

been

aware of the existence of fastidious

organisms

and have

compounded

enriched media for

growth

of such

organisms.

In addition,

specific

nutritionally deficient organisms such as

thiol-dependent

streptococci

and

Haemophilus

in-fluenzae

are commonly cultured in

symbiotic

culture. Inthis

study

wedescribe the isolation of

an unrelated

organism requiring

at least one

identifiedgrowth factor-folic acid. Unlikemany

of the

nutritionally

deficient

streptococci,

Pink-MDA cannot grow in the absence of the

nutri-tional

requirement

on subsequent culture. In

addition, Pink-MDA has growth characteristics similar to the

Legionella

species

in that most growth media

apparently

do not contain all of

the

necessary

nutritional

requirements

or else

containfactors that inhibit itsgrowth.

The discovery of organisms such as

Pink-MDA further demonstrates, as did the initial

.1

t

: i

f

on February 7, 2020 by guest

http://jcm.asm.org/

(6)

difficulty in growing the Legionella species, the

inadequacy of many primary isolation media.

The frequency of bacterial infections considered

viral because bacteria were not detected or

aseptic duetoadequate therapy thatwere

actu-ally caused by organisms similar to Pink-MDA

remains to be determined. We are, however,

presentlyinoculating all lung biopsy and aspirate

specimensontoboth folicacid-supplemented (10

,ug/ml) and nonsupplemented CYEA. To date,

Pink-MDA has notbeen isolated fromany

addi-tional specimens.

ACKNOWLEDGMENTS

The G+Canalysiswasperformed by Manley Mandel and

theelectron microscopywasdone by Michael Ahearn.

LITERATURE CITED

1. Barker, J., D. Healing, and J. G. P. Hutchison. 1972. Characteristics ofsomeco-trimoxazole-resistant Entero-bacteriaceae from infected patients. J. Clin. Pathol. 25:1086-1088.

2. Brenner, D. J., A. G. Steigerwalt, G. W. Gorman, R. E. Weaver, J. C. Feeley, L. G. Cordes, H. W. Wilkinson, C. Patton, B. M. Thomason, and K. R. Lewallen Sasseville. 1980. Legionella bozemaniisp. nov.andLegionella

di.-moffiisp. nov.:classification oftwoadditionalspecies of Legionella associated with human pneumonia. Curr. Mi-crobiol. 4:111-115.

3. Edelstein, P. H., and S. M. Finegold. 1979. Use of a

semiselectivemediumtoculture Legionella pneuimophila from contaminated lung specimens. J. Clin. Microbiol. 10:141-143.

4. Feeley, J. C., R. J. Gibson, G. W. Gorman, N. C. Lang-ford, J. K. Rasheed, D. C. Mackel, and W. B. Baine. 1979. Charcoal-yeastextractagar:primary isolation medium for Legionellapneumophila.J.Clin. Microbiol. 10:437-441. 5. Feeley, J. C., G. W. Gorman, R. E. Weaver, D. C.

Mackel, and H. W. Smith. 1978.Primary isolation media for Legionnaires disease bacterium. J. Clin. Microbiol. 8:320-325.

6. Garrity, G. M., A. Brown, and R. M. Vickers. 1980. Tatlockia andFluoribacter:twonew generaof organisms resemblingLegionellapneuimophila. Int. J. Sys. Bacteri-ol. 30:609-614.

7. Hartman, S. C. 1970. Purines andpyrimidines,p.50-51.

InD. M. Greenberg(ed.), Metabolic pathways, vol. IV, 3rd ed. Academic PressInc., New York.

8. Hebert, G. A.,A.G.Steigerwalt,and D.J.Brenner.1980.

Legionella micdadei speciesnova:classification ofathird

species ofLegionellaassociated with humanpneumonia. Curr. Microbiol. 3:225-257.

9. Lewallen, K. R., R. M. McKinney, D. J. Brenner, C. W. Moss, D. H. Dail, B.M. Thomason, and R. A. Bright. 1979. Anewly identified bacterium phenotypically

resem-bling, but genetically distinct from, Legioniella

pnieuimo-phila: an isolate in a case ofpneumonia. Ann. Intern. Med.91:831-834.

10. Mandel, M., E. R. Leadbetter, N. Pfennig, and H. G. Truper.1971. Deoxyribonucleic acid base compositions ofphototrophic bacteria. Int. J.Syst. Bacteriol. 21:222-230.

11. Mandel, M.,C. L. Schildkraut, and J. Marmur. 1968. Use ofCsCl density gradient analysis for determining the guanineplus cytosinecontent ofdeoxyribonucleic acid. Methods Enzymol. 12:184-195.

12. McCarthy, L. R., H. Chmel,G. Bell, and D. Armstrong. 1977. Thymidine-dependent strain of Sal,nonella oslo selected bytrimethoprim-sulfamethoxazole therapy. Am. J.Clin. Pathol. 68:307-311.

13. McKinney, R. M., R. K. Porschen, P. H. Edelstein, M. L. Bissett, P. P. Harris, S. P. Bondell, A. G. Steigerwalt, R. E. Weaver, M. E.Ein,D.S.Lindquist, R. S. Kops, and D. J. Brenner. 1981. Legionella longbeachae species

nova,anotheretiologicagentof humanpneumonia. Ann. Intern.Med. 94:739-743.

14. McKinney, R. M., B. M. Thomason, P. P. Harris, L. Thacker, K. R.Lewallen, H. W. Wilkinson, G. A. Hebert, andC. W. Moss. 1979.Recognitionofanewserogroupof Legionnaires disease bacterium. J. Clin. Microbiol. 9:103-107.

15. Morris, G. K., A. Steigerwalt, J. C. Feeley, E. S. Wong, W. T. Martin, C. M. Patton, and D. J. Brenner. 1980. Legionellagormaniisp. nov. J. Clin. Microbiol. 12:718-721.

16. Okubadejo, 0.A., and R. M. Maskell. 1973. Thymine

requiring mutants of Proteuis mirabilis selected by

co-trimoxazole in vivo. J. Gen. Microbiol. 77:533-535. 17. Pine, L., J. R. George, M. W. Reeves, and W. K. Harrell.

1979.Development ofachemically defined liquid medium

forgrowth of Legionellapneumophila.J. Clin. Microbiol. 9:615-626.

18. Ristroph, J. D., K. W. Hedlund, and R. G. Allen. 1980. Liquidmedium forgrowth of Legionella pneurnophila. J. Clin.Microbiol. 11:19-21.

19. Schildkraut,C. L., J. Marmur, and P. Doty. 1962. Deter-mination of the base composition ofdeoxyribonucleic acidfrom its buoyant densityinCsCl. J. Mol. Biol. 4:430-443.

20. Smalley, D. L., P. A. Jaquess, and J. S. Layne. 1980. Selenium-enriched medium forLegionella pneurnophila. J.Clin. Microbiol. 12:32-34.

21. Tapsall, J. W., E. Wilson, and J. Harper. 1974. Thymine dependentstrains of Escherichia coli selectedby trimeth-oprim-sulphamethoxazoletherapy.Pathology 6:161-167. 22. Warren, W. J., and R. D. Miller. 1979.Growth of

Legion-naires disease bacterium (Legionella pneiumophila) in chemically defined medium. J. Clin. Microbiol. 10:50-55.

VOL. 16, 1982