• No results found

Enzyme linked immunosorbent assay for antibodies against teichoic acid in patients with staphylococcal infections

N/A
N/A
Protected

Academic year: 2020

Share "Enzyme linked immunosorbent assay for antibodies against teichoic acid in patients with staphylococcal infections"

Copied!
7
0
0

Loading.... (view fulltext now)

Full text

(1)

Vol.17,No. 4 JOURNALOFCLINICAL MICROBIOLOGY, Apr. 1983,p.640-646

0095-1137/83/040640-07$02.00/0

Copyright©1983, AmericanSocietyforMicrobiology

Enzyme-Linked Immunosorbent

Assay for

Antibodies Against

Teichoic

Acid in

Patients

with

Staphylococcal Infections

MARTAGRANSTROM,'* INGERG.JULANDER,- SVEN-AKEHEDSTROM,3 ANDROLAND

MOLLBY'

Department ofBacteriology, NationalBacteriological Laboratory, S-105 21 Stockholm,' Department of Infectious Diseases, Roslagstull Hospital, KarolinskaInstitute, S-114 89Stockholm,2 andDepartment of

Infectious Diseases, University Hospital, S-223 62Lund,3Sweden

Received 26April 1982/Accepted 26 August 1982

A highly purified teichoic acid preparation was used in an enzyme-linked

immunosorbent assay to measure the specific immunoglobulin G (IgG) and IgM

response instaphylococcal disease. Antibody determination in a normal

popula-tion, showing a difference of up to 20-fold in the mean IgG titers between the

youngest children and adults, wasused toestablish age-correlated upper normal

values. IgM antibodies were found to be of little diagnostic value since their

response was often low or absent. Increased IgG titers were found in 24 of 27

(89%) patients withendocarditis, in 11 of14 (79%)with complicated septicemia,

and in 10of 20(50%) with uncomplicated septicemia with serum samples drawn

betweendays 7and 30 ofdisease. Withpaired samples, the numbers of patients

withincreased IgG titers were 17 of 17, 3of 4, and 6 of 7, respectively, in the same

patient groups. IncreasedIgGtiterswere lessoften demonstrated in patients with

chronic osteomyelitis (7 of 22). The enzyme-linked immunosorbent assay for

teichoic acid antibodies was found to be a sensitive and specific method for

diagnosing staphylococcal endocarditis andsepticemia. Foroptimal results, both

the substantial age-correlated variation in normal titers and the importance of

adequately spaced samples should be considered.

A reliable serological test for staphylococcal

infections is desirable for several clinical

pur-poses: to diagnose staphylococcal septicemiaor

endocarditis inpatientswithnegativeblood

cul-tures; to detect complications such as

endocar-ditis; andtodetect and followdeep

staphylococ-cal infections, e.g., osteomyelitis.

Thefirsttestavailable forroutine use wasthe

anti-staphylolysin test, but its value has been

considered limited because of its low sensitivity

(5). Antibodies against teichoic acid, acell wall

antigen of Staphylococcus auirelus, were first

demonstrated by Martin et al. (8) and by

Crowder and White (3) bygel immunodiffusion

in patients with various staphylococcal

infec-tions, especially endocarditis. The original

ob-servationsoftheseworkers wereconfirmed and

expandedinlaterworkbyseveralgroups.

Coun-terimmunoelectrophoresis wasfoundtobemore

sensitivethan immunodiffusion (6, 10, 14).

The radioimmunoassay has been found still

moresensitivethan

counterimmunoelectrophor-esis for detecting teichoic acid antibodies (18).

However, in general, the enzyme-linked

im-munosorbent assay (ELISA) is as sensitive as

the radioimmunoassay and more suitable for

routine laboratory work. It has been used to

determineantibodiestoteichoic acid inalimited

study by Mackowiak and Smith (7) and more

recently ina study by Verbrughet al. (15).

The aimof this research was to studytheuse

of the ELISA formeasuringimmunoglobulin G

(IgG) and IgM antibodies withapurifiedteichoic

acid preparation as antigen. Antibody levels

weredetermined inhealthy controls of different

ages as background for the levels in patients

with deep-seated or superficial staphylococcal

infections. The use of this test as an aid in

clinical diagnosis was evaluated.

MATERIALS ANDMETHODS

Serumsamples. Inall, 242 serumsamplescollected

from 140patientswith varioustypesofstaphylococcal

infections were tested. Inaddition,32 serum samples

from 20 patients with septicemia or endocarditis of

non-staphylococcal etiology and 169 serum samples

from healthycontrolsaged 0to94 years weretested.

Samples weretaken from patients admitted for

endo-carditis andsepticemia and from patients treatedfor furunculosis at the Roslagstull Hospital, Stockholm,

Sweden. Samples were also collected from patients

with chronic osteomyelitis and furunculosis at the Department for Infectious Diseases, University Hos-pital, Lund, Sweden.

Endocarditis (38 patients, 64 samples). Criteria for

640

on February 8, 2020 by guest

http://jcm.asm.org/

(2)

inclusion wereat leasttwobloodculturesyieldingS.

aureusand clinical features consistent with

endocardi-tis. The majority (26 patients) were narcotic drug

abuserswithseptic pulmonary embolism andtricuspid

valve endocarditis. In 10 of them, the diagnosiswas

verified byechocardiography. The remaining patients,

2 narcotic addicts and 10 nonaddicts, suffered from

endocarditis of the aorticormitral valvesdiagnosed by

recognition of significant murmurs orrevealed at

au-topsy.

Complicated septicemia (18 patients, 25 samples).

Criteria for inclusion in this groupwerepatients with

at least two positive blood cultures andclinical

evi-dence of septicemia with complicating foci. These

manifestations were osteomyelitis, arthritis, or

spon-dylitis in 10 patients, abscesses in 7 patients, and

purulentmeningitisandpneumonia in1 patient.

Uncomplicated septicemia (28 patients, 37samples).

Patients fulfilled the criteria for septicemia given

above but without detectablecomplicating foci.

Chronicosteomyelitis(22 patients,41 samples). The

diagnosiswasconfirmedby X-rayorhistopathological

examination ofbiopsies. Cultures yielding S. aureus

were performedrepeatedly from abscesses, deep

sam-ples from sinus tracts, and tissue biopsies. No other

bacteriawereobtained. The duration ofosteomyelitis

varied from0.5 to 67years.

Furunculosis(34 patients,43samples). Patients had

suffered fromatleast threeepisodes of furunculosison

theface, the trunk, or extremitiesduringaperiod from

2 monthsto 15 years.Samplingwasdoneduring both

the active and the inactive phases of the disease. S.

aureuswasrepeatedly cultured from the lesions.

Non-staphylococcal infections (20 patients, 32

sam-ples). Serumsamples were obtainedfrom 13 patients

withsepticemia and7withendocarditisof

non-staphy-lococcal etiology. In the cases of septicemia, the

etiologies were: Escherichia coli in seven,

alpha-he-molytic streptococci in four, beta-hemolytic

strepto-cocci infour, Staphylococcus hominis in one. In the

cases of endocarditis, alpha-hemolytic streptococci

wereresponsible in six and beta-hemolytic

streptococ-ciinone.

Preparation of teichoic acid antigen. Teichoic acid

waspurified as described by Baddiley and Davidson

(1). S. aureus strain Wood 46, which produces low

amounts ofprotein Aand contains aribitol teichoic

acid withp-linked N-acetylglucosamineresidues[type

A

(p)],

wasusedtoproduce the antigen. This type of

teichoic acid isreportedtobe presentin 96to100%oof

human isolates of S. aureus (9, 11). A 1-liter, 5-h

culture(caseinhydrolysate-yeastextractmedium)was

centrifuged (9,000 x g)at4°C, and the pelleted cells

were suspended in 0.1 M phosphate buffer, pH 7.2.

The bacteriawereheatedto70°C for10min,washed,

suspended in 50 ml ofbuffer, and disintegrated by

shakingwith glass beads for 30 min. After filtration

andcentrifugationat750 x g for30min,the

superna-tant was centrifuged at 16,000 x g for 10 min. The

pellet was washed twice in saline and suspended in

0.02 Mphosphate buffer, pH7.2. Thesuspensionwas

incubated with chymotrypsin (100 p.g/ml; Sigma

ChemicalCo., St. Louis, Mo.)and DNase(20,ug/ml;

Sigma) at 37°C for 120 min on a shaking table. The

pellet obtainedby centrifugation at 1,000 x g for10

minwasdiscarded, andthesupernatantwas

recentri-fugedat20,000x g for20min. Thispelletwas washed

once in saline, once inphosphate buffer, and twice in

water-saturated chloroform. The pellet thus obtained

was lyophilized and extracted three times with 10%

trichloroacetic acidfor72h at4°C. The solutionwas

centrifugedat10,000 x gfor10min, and the teichoic

acidwasprecipitatedoutof the supernatantby

incuba-tion withafourfold volume of absolute ethanolat0°C

for18 h. Theprecipitatewascollected by

centrifuga-tion, suspended in distilled water, and lyophilized.

This preparation yielded one strong line in crossed

immunoelectrophoresis (17) againstahigh-titer human

serum.

ELISA. The ELISA was done by the microplate

modification (16) of the method ofEngwall and

Perl-mann (4). Polystyrene microplates (M129 B;

Dyna-tech, Plochingen, West Germany) were coated by

incubationovernightatroomtemperaturewith100jil

of teichoic acid solution (1 ,ug/ml) in

phosphate-buff-eredsaline, pH7.2. Theplateswerewashed, and 100

,ul ofanappropriatepatientserumdilutionwasadded

toeach oftwocoated wells andtwouncoated wells.

The plates were then incubated for 1 h at room

temperature forIgG antibody determination and for2

h at 37°C for IgM determination. After washing the

plates, 100,ul of alkalinephosphatase-conjugated

anti-serumagainst humanIgG of IgM (OrionDiagnostica,

Helsinki, Finland) was added, and the plates were

incubatedovernightat room temperature. Afterfinal

washings, 100 ,ul of the substrate

p-nitrophenylphos-phate was added, and the plates were incubated at

room temperature for 60 min. All washings of the

platesweredone threetimes with phosphate-buffered

saline containing 0.05% (vol/vol) Tween 20. Reading

wasdoneautomatically in a Titertek Multiskan (Flow

Laboratories, Irvine, Scotland). Known positive and

negative controls were included in each test series,

and the resultswere correctedagainst these controls

tominimize day-to-day variation.

RESULTS

Optimum conditions of the test. The antigen

concentration necessary for coating the

micro-plates, as determined by serial dilution of the

antigen (Fig. 1), was 1 ,ug/ml. After overnight incubation, the platescould be emptied, covered

with Parafilm,and stored for 4 to 6 weeks at4°C without loss of antibody-binding ability.

End-point titrations of serum samples from patients

andcontrols indicatedabroadrange of titers. A

routine dilution of10,000 was optimal in most

cases for IgG antibody determinations. For a

few patients, mostly children, the optimal

dilu-tion was found to be 1:1,000, and for some

patients with exceptionally high levels of IgG

antibodies, a serum dilution of 1:100,000 was

needed. Theoptical densities of these dilutions

as measured by absorbance at 405 nm (A405)

correlated with the endpoint titers obtained by

serial 3.16-fold dilution (correlation coefficient,

0.98)(Fig.2). These values(i.e.,theA405 values

times thedilutionfactors)arereferred to

hereaf-ter as ELISA titers.

For IgM antibody determination, a serum

dilution of 1:1,000 was optimal in most cases.

on February 8, 2020 by guest

http://jcm.asm.org/

(3)

642 GRANSTROM ET AL.

5.0-E

c

U,

1.0-0

C

0 .00

O

0.1

1,000

500-0

C

0

._

0

c,

0

C,)

-w

-i

ul

I I I

0 1 ng 10ng 100ng Ipg Ogpg 100 g

Concentration of teichoic acid per ml

FIG. 1. Determination of optimal teichoic acid

antigen concentration for coating the microplates in

the ELISAasshownin arepresentative experiment.

Only intwocases was a1:10,000dilution

neces-sary. A significant rise of titerwasdefined as a

doubling ofthe titerfound in the previous

sam-ple.

Antibodies against teichoic acid in a normal

population.

Antibody

levels equal to those of

adults werefound in newborninfants. Antibody

levels decreased rapidly to minimum levels at

the age of6 months to 1 year (Fig. 3). These

antibodies were presumably ofmaternal origin

since the mothers had approximately the same

antibody titersatdelivery. Furthermore,in

con-secutive samples from five infants, a decrease

was found during the first half-year of life. Thereafter,acontinuous rise in titers was noted

up totheageof13 to 15years,when adultlevels

were reached.

Astheupperlimit ofnormalvalues, the mean

titerplustwo standarddeviationswas used (Fig.

3).However, becauseofthe skewdistribution of

the normal values, in the larger groupofyoung

adults (13 to 45 years) a titer of 8,000 to 9,000

was usedas acommonupper limit,

correspond-ingroughly to the 95th percentile. In all, 4% of

the healthy controls exceeded these upper

nor-mal limits. In the older age groups, decreasing

mean titerswere noted.

The background level of IgM antibodies was

alsoestablished. Various titers of low

nonspecif-ic IgM activity were found in all age groups.

Establishedupper limitsofnormal values varied

from 30 to 300 (children to 13 years) and from

200 to 450 inindividuals over 13years.

Nonspecific IgG activity varied from a mean

of40 in children to a meanof 50 in adults with

serum dilutions of 1:1,000.

Activity

was even

100-

50-

10-

5.0-0.1

ELISA

n

*.

*1*

0.5 1.0 5.0 10 50 100

IgG titer(A405 x dilution factor)

FIG. 2. Correlation between IgG antibody

end-point titers andrelative titers(A40Sxdilutionfactor)at

optimal serumdilution. Valuesonboth axes, x103.

lower with the routine dilution of1:10,000 used

for adults.Owing to these lowbackground

activ-ities, nonspecific binding was not subtracted

from the ELISAtiters.

Antibodies against teichoic acid in

staphylococ-calinfections. (i) Endocarditis. TheIgGandIgM

titers in all serumsamplesavailable, irrespective

of the time ofsampling,areshown inFig.4. The

titer values were judged positive or negative

according totheage-correlated limits described

above.HighIgGtiters(>20,000)weremarkedly

more common in the endocarditisgroup than for

other staphylococcal infections. Thehighest

ti-ters were recorded in drug addicts with

tri-cuspidal endocarditis. In four of these patients,

theIgG antibody titerexceeded100,000.

PositiveIgMtiterswereless oftenfound(in 11

of 64 serumsamples),and thetiterstendedtobe

low. In 10 of these 11 samples, the IgG titers

were alsohigh.

TheIgGantibodytiters inrelationtothe time

ofsampling are illustrated inFig. 5. Significant

rises intiters werefound in five patients. In all

these cases, the first sample was drawn before

day 11 of disease. Inserumsamplesdrawnatan

earlystage of disease(<7days), low titerswere

generally found. The

highest

titers were

ob-tained in samples drawn7 to 30

days

after the

onset of disease. The kinetics of

antibody

de-creasecould notbe studied in detailowingtothe

limitednumber of late samples. Titer decreases

were, however, noted afterday30in the

major-ityofpatientsfrom whom latefollow-up samples

wereavailable.

The IgG antibody responses in patients with

endocarditis are summarized in Table1. Out of

J.CLIN. MICROBIOL.

*0

on February 8, 2020 by guest

http://jcm.asm.org/

(4)

10,000

-C 5,000.

0

-J

a 1,000'

<. 500~

u 100.

co

LA 1-J

LU

V.

'0

,IA

,#

4. _

.I

,,

-' :/

0

0

0

0

A

0

I0

I

0

0

0

71

A

0

+0 .*

*

S

0

0

AL

0

0

0j

I1

0

0

0

0

0

0

I I I 1 I I ! - I

01-34-6 1 2-3 4-6 7-9 10-12 13-15 16-30 31-45 46-60 61-75 476

MONTHS YEARS AGE

FIG. 3. IgGantibodiesagainstteichoicacid in166healthycontrols of differentage groups.Dashed lineshows

meanvalues;solid bars showmeanvaluesplustwostandarddeviations,takenasupperlimits ofnormalvalues.

38 patients, 26 had positive titers (68%),

irre-spective of the time of sampling. However,

when the time ofsampling was taken into

con-sideration,itwasfound thatamongpatients with

serunm samples drawn between days 7 and 30

after the onset ofdisease, 24 of 27 (89%) had

positivetiters. Because of individual variations

IgG

0

x

'C

0

'C

-I w6

in the antibody response, adequate sampling

shouldincludetwosamples drawn before day 30

of the disease. Such paired sampleswere

avail-able from 17 patients, and all (100%) had

posi-tive IgG antibody titers (Table 1).

(ii) Complicated septicemia. TheIgG and IgM

titers in all serum samples available from

pa-1gM

m

rn

0 c

0 x

-1

FIG. 4. IgGandIgMantibodiesagainst teichoic acid in 242serumsamples from 140 patients withdifferent

staphylococcal infections. Symbols: *, titers above age-correlated uppernormal values (positive); 0, titers

belowage-correlateduppernormal values(negative).

on February 8, 2020 by guest

http://jcm.asm.org/

(5)

644 GRANSTROM ET AL.

ENDOCARDITIS

5 15 25 35 45 2 10

Days Months

COMPLICATED SEPTICEMIA

5 15 25 35 45 '2

Days Months

UNCOMPLICATEDSEPTICEMIA

t0

O

DWz

~~~~~

5 15 25 35 45 55

Days

Time after onset of illness

FIG. 5. IgG antibodies in patients with endocarditis and complicated and uncomplicated septicemia in relation todurationof illness.Symbols: 0,titers aboveage-correlateduppernormalvalues(positive);0,titers belowage-correlated uppernormal values(negative);solid linejoinsserum samplesfromsamepatient.

tients withcomplicated septicemiaareshown in

Fig. 4.Positive IgG titerswerefoundin13outof

18 patients. The highest IgG titers were

mea-sured in thetwodrugaddicts in thisgroup.The

other 11 patients with positive IgG titers were

nonaddicts, and none had a titer exceeding

20,000. Inpatientswithpairedormultipleserum

specimens, no significant rise in the IgG titer

was found (Fig. 5). Of patients with samples

drawn duringthe optimal sampling period (7to

30 days), 11 of 14 (79%) showed positive IgG

titers (Table 1).

(iii) Uncomplicated septicemia. Out of 28

pa-tients with uncomplicated septicemia, 12 had

increased IgG titers. Significant rises in IgG

titerswerefound in threepatients (Fig. 5). With

samplesdrawn betweendays7 and30, 10outof

20patients (50%)had positive IgG titers (Table

1). When paired samples were available, six of

sevenpatients were positive.

(iv) Chronic osteomyelitis. Few patients (8 of

22)with this disease hadpositiveIgGtiters(Fig.

4 and Table 1). Increased IgM titers were not

recorded (Fig. 4).

In six of eight patients, positive IgG titers

could be correlated with an active phase

(re-lapse) of chronic osteomyelitis, although many

relapses occurred without increases ofantibody

titers. Patients withanincreased IgGtiter had a

higher degree of soft tissue engagement, e.g.,

fistulae and abscess formations.

(v) Furunculosis. With this disease, few

posi-tive IgG and no positive IgM titers were

mea-sured (Fig. 4 and Table 1). No correlation

be-tween titers and the grade of activity of the

diseaseswas found.

(vi)Septicemia and endocarditisof

non-staphy-lococcal origin. Only one patient in this group

hadaslightly elevated IgG antibody titer (Table

1). This patient suffered from endocarditis

caused by alpha-hemolytic streptococci, and

there was no evidence of current or recent

staphylococcal infection. A frequency of1

pa-tient withpositivetitersoutof 20 (5%) is similar

to the frequency of positive titers in healthy

controls (7of166, or4.2%).

DISCUSSION

The ELISA was found to be a simple,

sensi-tive methodfor detecting specific antibody

re-sponses to the teichoic acid of staphylococci. The assay may be completed within 1 day by

decreasingtheincubation time withtheenzyme

conjugate.

The age distribution of IgG antibody levels

withincreasing titers from childhoodtoadult life

may indicate repeated antigen stimulation by

infection or colonization. This explanation

would also account for the low or absent IgM

antibodyresponserecorded inourpatients since

IgM antibodies after repeated exposure of an

antigen maybe atalow level or even

undetect-able(2). Similarobservations were reported by

Verbrugh etal. (15) withanothercell wall

com-ponent, thepeptidoglucan, usedastheantigen.

Rapiddisappearance of IgMantibodiescouldbe

an alternative explanation, but our data do not

support this hypothesis since increased IgM

titers were notmore commonin the veryearly

samples. Furthermore, even the specificity of

the few elevatedIgMtitersremainsquestionable

1

00,000-

50,000-1

0,000-

5'000-

1,000-

1-0

U a

c 0

5

0 ._G

x

0

IV

-2

._ w

J. CLIN. MICROBIOL.

0

0

0

0

0

0

---ve 0 0 0oe

0

on February 8, 2020 by guest

http://jcm.asm.org/

(6)

TABLE 1. ELISAIgGantibodytiters instaphylococcalinfections in relationtotime ofsamplingafteronset of illness

Positivesamples/Totalsamplesa

Typeof disease Allserumsamples Drawn7 to30days Early and

late'

Allserumsamples afteronset

Endocarditis 26/38 (68) 24/27(89) 17/17(100)

Complicatedsepticemia 13/18 (72) 11/14(79) 3/4 (75)

Uncomplicatedsepticemia 12/28 (43) 10/20(50) 6/7 (86)

Osteomyelitis chronica 8C/22 (36)

Furunculosis 6/34 (18)

Non-staphylococcal endocarditis 1/20 (5)

andsepticemia

None 7/166 (4)

aNumbers inparenthesesshowpercentagesofsamplesthatwerepositive.Forallstaphylococcal infections,

73% of7-to-13-day samplesand92% ofearlyandlatesampleswere positive.

bEarly samplesweredrawnbeforeday14ofdisease;late samples weredrawnbetweendays14and30of

disease.

cIncludingonepatientwithasignificantrise in titers.

because of rheumatoid factor, knownto appear

during the course of many infectious diseases

(12, 19).

Ourstudy showed the need for careful

popula-tion studies since large differences inmeantiters

wereobserved betweenchildren of differentage

groupsand adults(Fig. 3). Consequently,

sepa-rate cutoff levelsmust be established for

differ-ent age groups. Also, the time of sampling was

found to be critical for the evaluation of

anti-body response in patients. Repeated sampling

with at least one early sample (<14 days after

theonsetofdisease) andatleastonelatesample

(14to30days) improved thediagnostic value of

thetest considerably (Table 1).

Earlierreports(14, 18) have suggested that the

antibody response to teichoic acid might

dis-criminate betweenpatients with endocarditisor

complicated septicemia and those with

uncom-plicated diseases. However, our data do not

support this finding (Fig. 4). The difference

might result from the patient material

investigat-ed since all patients with titers above 20,000

without endocarditis were drug addicts. In the

endocarditisgroup,thehighest titerswerefound

inpatients withtricuspid valve involvement, but

all these patients were alsodrug addicts. Thus,

ourdatasupportthe observationofMackowiak

et al. (7) that narcotic drug abusers seem to

develop high titers early in thecourseof

staphy-lococcal septicemia, irrespective of

complica-tions. This difference inantibody response can

well be explained by earlier repeated S. aureus

infections knownto occurin drug abusers.

Dif-ferent antibody responses in the drug addicts,

who constituted a large proportion of the

pa-tients in this study, can also well explain the

small difference in titers between patients with

complicated and those with uncomplicated

sep-ticemia in our study group as compared with

earlier reports (18).

Ourresultsforpatientswithchronic

osteomy-elitis are in accord with earlier findings that

increased teichoicacid titers are seldom found in

thiscategoryofpatients (7, 13). Increasedtiters

were mostlyfound in patients with more

wide-spread relapses involving soft tissue

engage-ment. In asequestered infection, the amountof

teichoic acidmay notbehigh enoughtogive rise

to a substantial antibodyresponse.

In conclusion, the ELISA wasfound to be a

sensitive and specific method for the detection

of antibodies against staphylococcal teichoic

acid. With the establishment ofage-correlated normal controlvalues, and especially with

ade-quately spaced samples, the assay can be a

usefuldiagnostictoolforpatientswithsuspected septicemia with endocarditis of staphylococcal origin. Forthediagnosis of chronic sequestered infections, other antigenslikeexotoxins used in

the ELISA may beofgreaterdiagnosticvalue.

ACKNOWLEDGMENTS

We thank Eivor Pettersson for technical assistance and Elisabeth Hamlen forpreparing themanuscript.

This workwassupported bygrantsfromtheMarianne and MarcusWallenbergs Foundation and bygrant 16X-2562from theSwedish Medical Research Council.

LITERATURECITED

1. Baddiley, J., and A. L. Davidson. 1961.The occurrence and location ofteichoic acids in lactobacilli. J. Gen. Microbiol.24:295-299.

2. Cremer,N. E.1979.Antibody formationand theimmune response, p. 362. In B. A. Freeman(ed.), Burrows text-book ofmicrobiology,21st ed. The W. B.Saunders Co., Philadelphia,Pa..

3. Crowder, J. G.,and A. White.1972. Teichoicacid anti-bodies instaphylococcalandnon-staphylococcal endocar-ditis. Ann.Intern. Med. 77:87-90.

4. Engwall, E., and P. Perlmann. 1972. Enzyme-linked

on February 8, 2020 by guest

http://jcm.asm.org/

(7)

646 GRANSTROM ET AL.

munosorbent assay, ELISA. III. Quantitation of specific antibodies by enzyme-labeled anti-immunoglobulin in antigen-coated tubes.J.Immunol. 109:129-135. 5. Hedstrom,S.A. 1969. Generalandlocalantibiotic

treat-ment of chronic osteomyelitis. Scand. J. Infect. Dis. 1:175-180.

6. Jackson, L. J., M.I.Sottile, F. G.Aguilar-Torres, T.H. Dee,and M. W. Rytel. 1978. Correlationof antistaphylo-coccal antibody titers with severity of staphylococcal disease. Am.J. Med. 64:629-633.

7. Mackowiak,P.A., and J. W.Smith.1978.Teichoic acid antibodiesin chronic staphylococcal osteomyelitis.Ann. Intern. Med.89:494-496.

8. Martin, R.R., H.Daugharty,and A. White.1965. Staphy-lococcalantibodies andhypersensitivitytoteichoicacids in man,p. 91-96.Antimicrob. AgentsChemother. 1965. 9. Nagel, J.G., J. N. Sheagren, C.U. Tuazon, and T. A.

Cardella.1977.Teichoic acidsinpathogenic Staphylococ-cusaureus. J.Clin.Microbiol.6:233-237.

10. Nagel, J. G., C. U. Tuazon, T. A. Cardella, and J.N. Sheagren. 1975. Teichoic acid serologic diagnosis of staphylococcal endocarditis. Use of gel diffusion and counterimmunoelectrophoretic methods. Ann. Intern. Med. 82:13-17.

11. Oeding, P. 1973. Wall teichoic acids in animal Staphylo-coccus aureusstrains determined byprecipitation. Acta Pathol.Microbiol.Scand. Sect.B81:327-336.

12. Reimer, C. B.,C. M.Blank,L.C.Phillips, E.F.Logan Hunter, B.J. Pender, and B. E. McGrew. 1975. The

specificity of fetal IgM antibody oranti-antibody? Ann. N.Y. Acad. Sci.254:77-93.

13. Tuazon,C. U. 1981. Teichoic acid antibodyresponsein

patientswith S.aureusarthritisandosteomyelitis,p. 957-960. InJ.Jeljaszewicz (ed.), Staphylococci and

staphylo-coccalinfections.Zentralbi. Bakteriol.Suppl.10.Gustav FisherVerlagGmbH & Co.KG,Stuttgart.

14. Tuazon,C.U.,J.N.Sheagren,M. S.Choa,andD. Mar-cus. 1978. Staphylococcus aureus bacteremia:

relation-shipbetween formationofantibodiestoteichoic acidand

development of metastatic abscesses. J. Infect. Dis. 137:57-62.

15. Verbrugh, H.A.,R. Peters, M. Rosenberg-Arska, P.K. Peterson, andJ. Verhoef. 1981. Antibodies tocell wall

peptidoglucanofStaphylococcusauireusinpatientswith seriousstaphylococcal infections. J. Infect. Dis.144:1-9. 16. Voller,A., D. E.Bidwell,G.Huldt, and E. Engvall. 1974. A microplate method ofELISA and its application to

malaria. Bull. W.H.O.51:209-211.

17. Weeke,B.1973.Crossedimmunoelectrophoresis. Scand. J.Immunol.Suppl. 1:47-56.

18. Wheat, L.J., R.B. Kohler, andA. White. 1978.

Solid-phaseradioimmunoassay for immunoglobulinG

Staphylo-coccus aureusantibodyin seriousstaphylococcal infec-tion. Ann.Intern. Med.89:467-472.

19. Williams,R.C.,Jr., and H. G. Kunkel. 1962.Rheumatoid

factor, complement, and conglutinin aberrations in pa-tients with subacute bacterialendocarditis. J. Clin.

In-vest.41:666-675.

J.CLIN. MICROBIOL.

on February 8, 2020 by guest

http://jcm.asm.org/

References

Related documents

European Commission (2007), Joint report on Social Protection and Social Inclusion 2007, Directorate-General for Employment, Social Affairs and Equal Opportunities..

Conclusion- ASO positive patients suffered from Rheumatic fever showed significant improvement after receiving the Ayurvedic formulation ‘ Rasnadi Guggulu’ in this study..

Although the Watson estimate is not a n upper bound on the possible value of the true pricing error, the concurrence among all three measures i n this case and the fact that

In this paper, we present performance analysis of Dijkstra’s single source shortest path algorithm over multiple GPU devices in a single machine as well as

The Strata DK424i provides new system hardware architecture compatible and upgradeable to a future larger Strata system, using the same new Base and Expansion Cabinets.. These

arbitrary scheme of things in which three other people come in and separate us.” KERMODE : This makes it interesting I think to consider the place among your boys of

Internet of Things (IOT) is used to communicate with the devices. By using this devices could be controlled or monitored through the internet, IOT acts as a

For what concerns the heat exchangers the supplier of these components has furnished several data from whom it has been possible to draw the pressure drop curve as