INFECtION AND IMMUNITY, Feb. 1971, p. 342-349Copyright © 1971 American Society for Microbiology

Vol. 3, No. 2Printed in U.S.A.

Cecal Enlargement and Microbial Flora in SucklingMice Given Antibacterial DrugsDWAYNE C. SAVAGE AND JULIA SUE McALLISTER

Department of Microbiology, University of Texas, Austin, Texas 78712

Received for publication 28 September 1970

Enlargement and microbial colonization of the cecum were examined in neo-

natal mice suckling mothers drinking either water or an aqueous solution of peni-cillin. The full ceca increased in weight at the same rate in both drug-treatedand control mice during the first 15 to 17 days after birth. Thereafter, cecal weightincreased at a greater rate in the drug-treated animals than in the untreated con-trols. At weaning, the ceca in treated mice were two to three times the size ofcontrol organs and remained enlarged as long as penicillin was given. The en-larged ceca did not differ histologically from those in controls. From birth,the cecal microflora in the drug-treated mice differed qualitatively and quantita-tively and in colonization pattern from the flora of control mice. The ceca ofuntreated animals were colonized primarily by large populations of lactobacilliduring the first week after birth, small populations of coliforms and enterococciduring the second week, and enormous populations of bacteroides and certain gram-negative fusiform-shaped anaerobic bacteria during the third week. In contrast,the organs of the treated mice were populated by large populations of coliformsand enterococci during the first week and enormous populations of clostridia andunusual gram-negative nonsporeforming bacteria during the third week. Theselarge abnormal populations were present in the ceca as they enlarged duringthe third week after birth in the drug-treated animals. These findings confirmthat only certain populations of anaerobic bacteria can act to maintain cecal sizein normal animals.

When adult mice are given solutions of certainantibacterial drugs for drinking water, theirceca fill with fluid and enlarge three to fourtimes normal size within 12 to 18 hr after thetreatment is started (10). Depending upon thedrug used, the enlarged ceca lose all or most oftheir indigenous microbial flora during the first24 to 48 hr but are colonized later by a microfloracharacteristic of the particular drug (2, 10). Forthat first 24 to 48 hr of treatment, however, thececa of these mice markedly resemble theenlarged ceca of germ-free rodents (10).

Germ-free rodents have large ceca, in partat least, because their ceca accumulate a sub-stance that induces smooth muscles to relax (4).Apparently, this substance is synthesized con-tinuously in the small bowel of all rodents but isinactivated in the cecum by the indigenous micro-flora in conventional animals (4).Such a substance may induce the enlarged

ceca of mice given antibacterial drugs. How-ever, once the ceca of drug-treated mice en-large, they remain large as long as the drug treat-

ment is continued even though they are recolo-nized by enormous microbial populations (10).Presumably, such an extensive microbial florawould inactivate the substance aHle to inducemuscle relaxation (4).These considerations make necessary alterna-

tive, but not necessarily mutually exclusive,hypotheses to explain the enlarged ceca indrug-treated mice. One such alternative hypothe-sis is that some indigenous microorganisms mayplay an essential role in maintaining the integrityof the water-transport mechanism in the intestinalepithelium (6, 10). Water accumulates in thececal lumen in mice given antibacterial drugs intheir drinking water for the first 24 to 48 hr of thetreatment (7, 10). Indeed, during that period, thececa increase in weight almost entirely becausewater collects in their lumens (10). The water mayaccumulate because normal water passage in andout of the ceca may be influenced in some wayby the indigenous microbial flora (6, 10).The microorganisms most likely to be involved

in any such influence on cecal physiology are the342

on October 19, 2017 by guest

http://iai.asm.org/

Dow

nloaded from

CECAL ENLARGEMENT AND MICROBIAL FLORA

anaerobic bacteria that predominate in the cecalmicroflora (7, 10). Particular anaerobic bacteriacolonize the intestinal tracts of mice during thesecond to third week after birth (11, 13). If suchbacteria are prevented from colonizing normally,ceca of baby mice should enlarge during thatperiod.We have tested this hypothesis in baby mice

suckling mothers drinking penicillin solution. Inthis report, we show that the ceca do enlarge insuch drug-treated neonates during the second orthird week after birth. The enlarging ceca arecolonized by large populations of microorganismsof various types including anaerobic bacteria.These findings and their implications are detailedin the following paragraphs.

MATERIALS AND METHODSMice. Swiss mice were obtained from the specific

pathogen-free (SPF) NCS-D (8) colony maintainedat the Rockefeller University (New York, N.Y.), theSPF Ha/ICR colony of A. R. Schmidt (Madison,Wis.), and the conventional colony of Euer's and Sons(Austin, Tex.).Gravid females from the various colonies were

maintained in individual cages with paper tops (Iso-cage, Bioquest) containing wood shavings for bedding.The females were given commercial pellets and acidi-fied water (1). The date of birth of the young wascarefully recorded so that the experiments could beconducted with animals of precisely known age.

Antibacterial drugs. Within 24 hr after delivery oftheir young, mother mice were given aqueous solu-tions of penicillin for drinking water (see experi-ments) .

Preparation of ceca for bacteriological examina-tion. Animals were sacrificed under chloroformanesthesia and weighed. Their ceca, always withcontents intact, were weighed and then homogenizedin a Teflon grinder in 5 ml of sterile charcoal-water(12).

Bacteriological culture techniques. The homogenatesdescribed above were diluted in charcoal-water in 10-fold steps. Calibrated loopsful of each dilution werethen spread on the surface of various selective agarmedia. The selective media and conditions of incuba-tion used for the recovery and enumeration of lacto-bacilli, coliforms, enterococci, bacteroides, andclostridia have been described (12, 13). The method ofestimating the populations of oxygen-intolerantanaerobes was developed from previously describedsystems (5). A vinyl chamber was modified from astandard vinyl isolator intended for housing germ-free animals (Germ-free Equipment, Palatine, Ill.).Bacteriological work was carried out in this chamberafter it had been exhausted by vacuum and refilled atleast five times with deoxygenated nitrogen. After thistreatment, methylene blue indicator remained color-less in the chamber for from 1 to 2 hours. The de-oxygenated nitrogen was prepared by passing com-mercial oil-pumped nitrogen over heated copper in agas purifying furnace (Sargent).

The bacteriological medium 0 and D (9) used forculturing oxygen-intolerant anaerobes was supple-mented with 5% human blood. We deoxygenated themedium prior to use in the chamber by incubating itfor at least 3 days at 37 C in an atmosphere of carbondioxide and nitrogen (13). All media and other mate-rials including dead mice were introduced into thechamber via an airlock that could be exhausted re-peatedly by vacuum and refilled with deoxygenatednitrogen. The mice were killed with chloroform justprior to being put into the chamber. The bacteriologi-cal methods were carried out in the chamber as de-scribed (5, 12, 13), except that charcoal-water wasused only after it had been boiled and then in-cubated for at least 3 days at 37 C in an atmosphereof CO2 and N2 (13). All media inoculated in thechamber were incubated in anaerobic jars in a H2-CO2 atmosphere (GasPak, Bioquest).

Histological techniques. The ceca with contentsintact were frozen in a 2% solution of methyl cellu-lose (15 centipoises) in 0.15 M saline on the freezingshelf of a microtome-cryostat (International Equip-ment Co., Needham Heights, Mass.). Sections of thefrozen tissues were cut at either 4 or 8 ,um, fixed for60 sec in absolute methyl alcohol, and stained withhematoxylin and eosin or a modified tissue Gramstain (11).

RESULTSBody weights of neonatal mice suckling mothers

drinking penicillin solution. On the day of deliveryof their young, mother mice from the SPF NCS-Dand Ha/ICR and the conventional Euer's colonieswere given penicillin solution (0.3 g/liter) todrink. The drug treatment was continued fromthat time until 5 weeks after delivery. The peni-cillin solution was renewed daily. At intervalsafter birth, babies from the treated group andfrom a control group drinking acidified waterwere killed and weighed. Their ceca were thenremoved and either weighed, cultured for bac-terial flora, or examined histologically as isdescribed in subsequent sections.The penicillin in the maternal drinking water

did not affect appreciably the gain in body weightof the NCS-D and the Ha/ICR neonates frombirth to weaning (Table 1). In contrast, the Euer'sneonates suckling mothers drinking penicillinsolution gained weight somewhat more rapidlythan their control counterpacts. However, in boththe treated and control groups, Euer's mice grewmore slowly and attained lower weaning weightsthan did animals from the two SPF colonies.Such observations are made frequently whensimilar experiments are carried out with mice fromcolonies housed conventionally (3).

Cecal weights in neonatal mice suckling mothersdrinking penicillin solution. In some of the experi-ments described above, weights were made ofceca removed with contents intact at intervals

'VOL. 3,2 1971 343

on October 19, 2017 by guest

http://iai.asm.org/

Dow

nloaded from

SAVAGE AND McALLISTER

Cs 1% 0 I-W) o-

. .I

00 N N 2IZ00C1 Ci 4I-- - . c-.cN 0 0 ON- ci c

~~~*00 00 00

NC,0~ ~~ ~ON 'N

00 -

00 " 'In " CI0

O) CO O 1_ - _ I _-q(: t- o _ z

0- .0 ON

0-8 I- X- I--

* NO £C -** I"v* *- * CN*

_. O 1 ON oo ON OI NI

ON 00 00 00 tn ~

0 * N tntcI...Ot ON * o00 oo

^')ON *ON *ON en N

00 I -N "N-j -2

00 cNE 00 0 N fICON 0 00o 00 -00o *ON> *00o *ON N * C- ~-I 001 00 £I O

N 00\0 001- IN ZZ tn W

N oC NNo N

Ct":T00 'IT00 ZON Cs tl Z00 li WIN. . .NFC Nrt S OO N- N oo 0r

Z 00 ON ri kn C

-c%C W) 'l tnWN mCON t W

IN IN tNt

ON 00

In

e

ct Ccn en r

00 en oN Cs tIN

*I

.i C

en en c

b,O a) IC ) CL)

m oC_ 4V. C ap cu ~ C.) C)

a. 04 a.

= C

0 aac

0>C2 co- C/-

C.d C)Z z z X u LU

344

0

0

cO

INFEC. IMMUN.

0

c0

C)O

0CO

0

-_

0

CO

z

C)

C)

OOCO

C.

CO

CO

2. ~

so-

oq

-*_

-

*_

o

-

*_

*_

co

s*_

Q)

0-*_

-*_

sv

o

a*_

-

q

Cq

-

o--

Cq

o

-

.R

o

-*

q

q

ss

-

-

a:

F

C)

on October 19, 2017 by guest

http://iai.asm.org/

Dow

nloaded from

VOL. 3, 1971 CECAL ENLARGEMENT AND MICROBIAL FLORA 345

TABLE 2. Increases in cecum weight in mouse neonates suckling mothers drinking either acidified water orpenicillin solution

Mouse type Drug

NCS-D Water5

NCS-D

Ha/ICR

Ha/ICR

Euers

Euers

Penc

Waterb

Penc

Waterb

Penc

Age of mice in days

5

(2-9)

(3-7)

4(3-5)

4(2-5)

(2-8)

3(2-4)

7

12(9-15)

11(9-13)

6(4-9)

S(4-8)

3(2-5)

5(4-5)

9

13(11-19)

17(14-18)

9(7-11 )

10(8-14)

5(2-9)

5(4-12)

11

15(11-20)

33(25-50)

14(11-17)

11(10-14)

16(13-18)

13(7-16)

13

24(19-29)

52(39-70)

25(21-29)

21(12-28)

18(17-31)

27(12-38)

15

34(30-39)

58(51-60)

30(18-37)

33(22-37)

18(12-24)

20(20-29)

17

40(31-43)

91(82-100)

41(32-67)

50(46-81)

30(23-34)

45(28-49)

19

130(101-185)

158(110-235)

143(112-160)

242(202-361 )

ND

ND

a Cecum weight in mg; median, range in parentheses; ND, not done; 15 to 20 animals per group.

b 0.001 N HCI.c A 0.3-g amount of penicillin per liter of water.

21

167(101-190)

367(252-435)

199(144-229)

315(255-331 )

108(78-143)

247(232-262)

after birth from neonates suckling mothers drink-ing either the penicillin solution (0.3 g/liter)or acidified water (Table 2). The full ceca en-larged at approximately the same rate in the con-trol and treated babies in the Ha/ICR and Euer'sgroups for the first 15 to 17 days. Thereafter, inthose mice cecum size increased most rapidly inbabies suckling mothers drinking penicillin solu-tion. In fact, the ceca enlarged much morerapidly in the treated than in the control animals;by 30 days after birth the organs of the drug-treated animals were two to three times the size ofthe control ceca. The NCS-D babies sucklingmothers given penicillin were somewhat an excep-tion to this general pattern; in these mice the fullceca began to increase in size at greater thancontrol rates at about 11 to 13 days after birth.In other respects, however, the enlargement proc-ess in the NCS-D neonates was identical to thatseen in the other types of mice.

Colonization by aerobic and microaerophilicbacteria of the ceca of neonatal mice. Culturesfor aerobic, facultatively aerobic, and micro-aerophilic bacteria were made from some of thececa removed at various intervals after birthfrom the neonatal mice suckling mothers drinkingeither acidified water or penicillin solution (Table3). Only mice from the two SPF colonies were

used in these experiments. The Euer's mice were

omitted because they were conventional animalswith typically complex populations of indigenousmicroorganisms (12).The bacteria able to grow aerobically or micro-

aerophilically colonized the ceca of the control

babies from both SPF colonies in much the same

fashion as has been reported for NCS and NCS-Dmice (11, 13). Populations of lactobacilli ap-

peared soon after birth and rose to the high levelscharacteristic of adults within the first week afterbirth. Populations of coliforms and enterococciappeared in the second week after birth, increasedquickly to substantial levels, and then declinedsomewhat to the lower levels characteristic ofadults during the third week after birth.

In contrast, these facultative and microaero-philic bacteria colonized the ceca of the drug-treated babies in quite a different pattern. Lacto-bacilli could not be cultured at any time afterbirth from the neonates suckling mothers drink-ing penicillin. Strikingly, however, populations ofcoliforms and to some extent enterococci estab-lished within the first week after birth increasedquickly to quite high levels and remained at thoselevels into adulthood. Such high population levelsof coliforms and enterococci are seen also inadult mice given penicillin solution to drink(2, 10).

Colonization by anaerobic bacteria of the ceca

of neonatal mice. Cultures for anaerobic bacteriawere also made from some of the ceca removedfrom the neonatal mice (Table 4). Anaerobestolerant of oxygen were grown in an atmosphereof CO2 and N2 (11) on -S medium (13) inocu-lated in the air. Such organisms were culturedfrom both NCS-D and Ha/ICR neonates.Anaerobes intolerant of oxygen were grown in anH2-CO2 (Gas Pak) atmosphere on deoxygenated

30

380(325-435)

850(681-944)

369(321-435)

1042(850-1190)

ND

ND

on October 19, 2017 by guest

http://iai.asm.org/

Dow

nloaded from

SAVAGE AND McALLISTER

TABLE 3. Colontizationz by aerobic aiid microaerophilic bacteria of the ceca ofNCS-D antd Ha/ICR mousenzeoniates sicklinig mothers drinlkinig either acidified water or penlicillini solutiosla

Mice

NCS-D

NCS-D

NCS-D

NCS-D

NCS-D

NCS-D

Ha/ICR

Ha/ICR

Ha/ICR

Ha/ICR

Ha/ICR

Ha/ICR

Drug

Waterb

Water

Water

Pend

Pen

Pen

Water

Water

Water

Pen

Pen

Pen

Bacteria

Lactobacillie

Coliformsc

Enterococci

Lactobacillic

Coliformsc

Enterococci

Lactobacilli

Coliformsc

Enterococci

Lactobacilli

Coliformsc

Enterococci

Age of mice in days

5

4

N

N

N

N

N

4

N

7

N

7 9

9 9

N

N

N

6

N

8

N

7

4+

N

N

9

N

8

N

N

N

9

4

11

9

6

N

9

9

8

4

N

10

8

13

9

6

7

N

9

10

8

4

4

N

10

7

15 17

9 9

6 5

6

N

10

10

8

4

6

N

10

9

4

N

10

10

8

4

6

N

10

9

19 21

9 9

4

4

N

10

10

8

6

4

N

10

10

4

4

N

10

10

8

4

4

N

10

9

30

9

4

4

N

10

10

8

4

4

N

10

9

a The data are recorded as the average of the loglo of the numbers of bacteria per gram of fresh tissue;15 to 20 animals per group. The i indicates bacteria infrequently cultured in low numbers; N, no bac-teria recovered.

b 0.001 N HCI.c The coliforms are predominantly slow fermenters of lactose (SLF) in untreated NCS-D and rapid

lactose fermenters in untreated Ha/ICR mice. In penicillin-treated animals of both mouse strains thecoliforms are either Klebsiella sp. or rapid lactose fermenters that form mucoid colonies.

d A 0.3-g amount of penicillin per liter of water.e In NCS-D mice the designation lactobacilli includes anaerobic streptococci (Gr.N) as well as Lac-

tobacillus sp.

O and D medium (9) supplemented with 5X%human blood and inoculated in an atmosphere ofdeoxygenated N2. Cultures for these oxygen-intolerant bacteria were made only from Ha/ICRinfants.The anaerobes tolerant of oxygen and able to

grow on -S medium colonized the ceca of thecontrol babies from both mouse colonies in apattern essentially identical to that previouslyreported for NCS and NCS-D mice (11, 13).These bacteria are mostly bacteroides; theycolonize the infant cecum at the end of the secondweek after birth and quickly establish high popu-lation levels characteristic of the adult.

Penicillin treatment delayed somewhat thecolonization of the cecum by oxygen-tolerantanaerobes but did not reduce the final populationlevels of such bacteria. In fact, penicillin-treatedHa/ICR babies yielded from their ceca higherpopulation levels of these anaerobes than did

untreated babies. More importantly, however,the populations of such bacteria in these drug-treated animals consisted mostly of clostridiarather than bacteroides. Therefore, similarly toits effect on the facultative and microaerophilicflora, penicillin induced marked changes in thequality of this particular fraction of the anaerobicflora.The fraction of the anaerobic microflora that

is intolerant of oxygen was cultured on 0 and Dmedium supplemented with human blood. Thismedium is not selective for particular types ofintestinal bacteria. Consequently, when incu-bated anaerobically, it will support the growth offacultatively anaerobic as well as strictly anaer-obic bacteria. Therefore, this medium must beinoculated with cecal homogenates both in theair and in a deoxygenated N2 atmosphere andthen incubated anaerobically. This procedurepermits more accurate estimates to be made of the

346 INFEC. IMMUN.

on October 19, 2017 by guest

http://iai.asm.org/

Dow

nloaded from

CECAL ENLARGEMENT AND MICROBIAL FLORA 347

TABLE 4. Coloniizationi by anaerobic bacteria of the ceca ofNCS-D anzd Ha/lICR mouse neontates sucklingmothers diriniking either acidified water or penicillin solhutionla

Drug

WatercPend

WatercPend

WatercPend

WatercPend

Age of mice in days

5

NN

NN

77

7;7

7

NN

NN

7

7

87

9

NN

NN

67

97

11

NN

N

N

ND8

98

13

8N

89

98

i 15 1 X 19

8N

8

99

1010

21 30

10 10 104 10 10

8 8 84 10 10

8 ND 89 ND 10

11 ND 1010 ND 10

1010

810

810

1010

a Data recorded as the average of the log10 of the numbers of bacteria per gram of fresh tissue; ±,

bacteria infrequently cultured in low numbers; ND, not done; 15 to 20 animals per group.b Anaerobes tolerant of oxygen are cultured in an atmosphere of CO2 and N2 on media inoculated in

the air. Under these conditions bacteroides are the most numerous organisms cultured on - S medium(13) from untreated mice of both colonies. In contrast, such cultures yield clostridia almost exclusivelyfrom mice treated with penicillin. Anaerobes intolerant of oxygen will grow in an H2-CO2 (GasPak)atmosphere on prereduced media inoculated in an atmosphere of deoxygenated nitrogen (5). Underthese conditions, fusiform-shaped and spiral-shaped microorganisms (5) from very high dilutions ofsample material and lactobacilli, coliforms, and bacteroides from lower dilutions can be cultured fromuntreated animals on 0 and D medium.

C 0.001 N HCl.d A 0.3-g amount of penicillin per liter of water.

4

* St

4

.I

I*e v +.5)X

I'

4 .|

FIG. 1. Gram-stailed histological sectionis of the ceca ef 17-day-old Ha ICR mice suhcklilng mothers drilik-inig either: (a) acidified water (0.001 N HCI) or (b) aquteouis peniicillini solution (0.3 glliter). Sectionis showv mut-cosal epitheliuim at the bottonm, mutcous layers on1 the epithelium, anid mixed microbial popuilatiolns at the top. Fatsi-f'rm-shaped bacteria cani be seeni in the mucous layer in (a). X 2,850.

populations of oxygen-intolerant anaerobes. Theestimates made from counts of colonies gcown on

plates inoculated in air can be subtracted from theestimates made from counts of colonies grown onplates inoculated in N2 (Table 4).

The results of this manipulation show thatpopulations of oxygen-intolerant anaerobes pre-dominate over other bacterial populations inuntreated babies over 13 days of age. The manipu-lation does not permit, however, the same con-

VOL. 3, 1971

MIice

NCS-DNCS-D

Ha/ICRHa/ICR

Ha/ICRHa/ICR

Ha/ICRHa/ICR

Types ofbanaerobes

02=tolerant

02=intolerant

02=tolerant

02=intolerant

Medium

-S-s

-s-s

0 and DO and D

O and DO and D

on October 19, 2017 by guest

http://iai.asm.org/

Dow

nloaded from

SAVAGE AND McALLISTER

clusion' for treated infants. In these animals,counts of coliforms and oxygen-tolerant anaer-obes are much the same as the counts of oxygen-intolerant anaerobes. Moreover, similar types ofmicroorganisms, i.e., clostridia and long, thingram-negative rods, grow on the 0 and D mediainoculated either in the air or in deoxygenated N2with cecal material from treated animals. Incontrast, in untreated mice the oxygen-intolerantpopulations consist predominantly of fusiform-shaped bacteria. Consequently, the anaerobicpopulations in the treated mice, though enor-mous, differ considerably from the analogouspopulation in untreated animals.

Histology of the cecal microflora in neonatalmice. Frozen-section histological preparationswere made of many of the ceca removed fromthe neonatal mice suckling mothers drinkingeither acidified water or penicillin solution (Fig.1). The sections shown are from ceca of 17-day-old mice. At this time populations of anaer-obic bacteria predominate in the ceca of boththe treated and untreated mice. However, asindicated by culture results fusiform-shapedbacteria predominate in the lumen and on themucosal epithelium in the sections of the cecaremoved from the control babies. In contrast,coccal-shaped gram-positive and long, thin,rod-shaped, gram-negative microorganisms pre-dominate in the sections of the organs fromtreated mice. Importantly, these microorganismsdo not populate in layers the epithelial mucinas do certain fusiform-shaped bacteria in un-treated animals. These histological findingsconfirm the culture findings; i.e., the predominat-ing types of microbial species in the cecal micro-flora of treated animals differ markedly fromthose in the flora of untreated mice.

DISCUSSIONIn this study we show that baby mice suckling

mothers drinking penicillin solution experiencececal enlargement during the second or third weekafter birth. Depending upon the type of micetested, this period coincides with or followsimmediately after the time when strictly anaerobicmicroorganisms colonize the large bowel innormal untreated mice (11, 13). Thus, the cecaenlarge only during or after this anaerobic micro-bial flora is disrupted in the penicillin-treatedmice. This timing may be only coincidental,however; the treated animals may experiencececal enlargement during the second or thirdweek after birth only because at that time theytake in more penicillin than they had previously.Neonatal mice begin to sample water and solidfood during the second to third week after birth

(personal observation). Consequently, the treatedbabies may take in more penicillin by drinking thedrug solution than they receive from theirmothers' milk. Thus, the ceca may enlarge attwo to three weeks of age in the treated babiesbecause only at that time is sufficient penicillinpresent in their guts to be effective in altering theflora.There is no doubt, however, that at the dosage

levels used sufficient penicillin passes through themother's milk to disrupt considerably the micro-bial flora from birth onwards. Lactobacillicolonize during the first week after birth inuntreated mice but never colonize at all in thetreated animals. In contrast, coliforms colonizeand reach high populations during the first weekin the treated mice but not until the second weekin untreated controls. These alterations in themicroflora during the first week do not lead toimmediate enlargement of the ceca, however;such enlargement begins to take place only at theend of the second week. Consequently, cecal en-largement in the drug-treated babies is not duesimply to alteration of the facultatively aerobicand microaerophilic fraction of that microflora.The organs enlarge only after alteration of the an-aerobic fraction as well. As previous findings haveindicated (6, 10) elements of the anaerobicmicroflora in adult mice may play a role in main-taining normal cecal size. Therefore, alteration ofthe anaerobic microflora is probably the keyelement in enlargement of the ceca in neonatalmice.

In this regard, it is important to stress that onlyalteration and not complete destruction of themicroflora is necessary for the ceca to enlarge(10). Many of the enlarging organs contain anextensive abnormal microbial flora includingenormous populations of anaerobic micro-organisms. Large populations of coliforms,enterococci, clostridia, and unusual nonspore-forming gram-negative bacteria can be found inthe ceca of the mice treated with penicillin.In contrast, only small populations of coliformsand enterococci, large populations of lactobacilliand bacteroides, and enormous populations ofgram-negative fusiform-shaped bacteria can befound in the ceca of the untreated mice (11,13).

In these untreated animals the fusiform-shapedanaerobes colonize in layers the mucin on theepithelium of the cecum and colon during thethird week after birth (11). These layers do notdevelop in the baby mice treated with penicillin.Similarly, these particular microbial layers dis-appear from the ceca of mice treated as adultswith penicillin and other antibacterial drugs (10).

348 INFEC. IMMUN.

on October 19, 2017 by guest

http://iai.asm.org/

Dow

nloaded from

CECAL ENLARGEMENT AND MICROBIAL FLORA

They do not reappear in such animals given thedrug for a sustained period even though thececa may be colonized thereafter by enormousmicrobial populations. Therefore, the gram-

negative oxygen-intolerant fusiform-shaped bac-teria that form the layers on the bowel epitheliumseem certainly to be involved in maintainingnormal cecal size (10).The mechanism of such a physiological inter-

action remains obscure. The microorganisms mayact to enhance motility of the bowel by destroyinga substance that induces relaxation of smoothmuscle (4). They may act as well to influence thewater transport mechanisms of the cecal mucosa(6, 10). We believe that studies of these problemsshould include careful examination of the rolesof the oxygen-intolerant anaerobic bacteria.

ACKNOWLEDGMENTS

This investigation was supported by Public Health Serviceresearch grant AI-08254 from the National Institute of Allergyand Infectious Diseases. The work with NCS-D mice was accom-

plished while the senior author was a Guest Investigator withRene Dubos at the Rockefeller University in New York, N.Y

LITERATURE CITED

1. Dubos, R. J., and R. W. Schaedler. 1960. The effect of theintestinal flora on the growth rate of mice and on theirsusceptibility to experimental infections. J. Exp. Med. 111:

407-417.

2. Dubos, R., R. W. Schaedler and M. Stephens. 1963. Theeffect of antibacterial drugs on the fecal flora of mice.J. Exp. Med. 117:231-243.

3. Dubos, R., R. W. Schaedler, and R. Costello. 1963. Theeffect of antibacterial drugs on the body weight of mice.J. Exp. Med. 117:245-254.

4. Gordon, H. A. 1965. Demonstration of a bioactive substancein caecal contents of germfree animals. Nature (London)205:571-572.

5. Lee, A., J. Gordon, and R. Dubos. 1968. Enumeration of theoxygen sensitive bacteria usually present in the intestinesof healthy mice. Nature (London) 220:1137-1139.

6. Loeschke, K., and H. A. Gordon. 1970. Water movementacross the cecal wall of the germfree rat. Proc. Soc. Exp.Biol. Med. 133:1217-1222.

7. Meynell, G. G. 1963. Antibacterial mechanisms of the mousegut. II. The role of Eh and volatile fatty acids in the normalgut. Brit. J. Exp. Pathol. 44:209-219.

8. Mushin, R., and R. Dubos. 1965. Colonization of the mouseintestine with Escherichia coli. J. Exp. Med. 122:745-757.

9. Omata, R. R., and M. N. Disraely. 1956. A selective mediumfor oral fusobacteria. J. Bacteriol. 72:677-680.

10. Savage, D. C., and R. Dubos. 1968. Alterations in the mousececum and its flora produced by antibacterial drugs. J.Exp. Med. 128:97-110.

11. Savage, D. C., R. Dubos, and R. W. Schaedler. 1967. Thegastrointestinal epithelium and its autochthonous bac-terial flora. J. Exp. Med. 127:67-76.

12. Schaedler, R. W., and R. J. Dubos. 1962 The fecal flora ofvarious strains of mice. Its bearing on their susceptibilityto endotoxin. J. Exp. Med. 115:1149-1160.

13. Schaedler, R. W., R. Dubos, and R. Costello. 1965. The de-velopment of the bacterial flora in the gastrointestinaltract of mice. J. Exp. Med. 122:59-66.

349VOL. 3, 1971

on October 19, 2017 by guest

http://iai.asm.org/

Dow

nloaded from