Table Of Content1@s-27-1995 15:3_ BIO-5 4@7 853-422@ CCAFS P.01
NASA.CR_203744 / [:;/_..'./<_,..;__)',
, d-V&
164 L#borator¥Aniolf#_(19_3)27,164-170
Comparison of garage, water bottle, and a high-moisture
diet bolus as dosing methods for quantitative D-xylose
administration to B6D2FI (Mus musculus) mice
J. PAUL ZIMMER t, SHERRY M. LEWIS z &JERRY L, MOYER 3
tDivi$1oo of Nutritional Sciences, Cornea UtQverslty, Ithaca, NY 148_0; aThe Bionctlcs Corporation, National
Center/or Toxicological l_esearch, Jefferson, All 72079; _nd JThe B/onettcs Corporattor6 NA,$A, Kennedy
Space Center, FL 32899, USA
Keywords: Dosing methods; Highomoisture diet;
Summtwy
Garage, water bottle, and diet lncorporstlon Meal; Garage; Water bottle; D-xylose; Xylose
are 3 do_lng methods used orally to administer tolerance test
test compounds to rodents, The_e 3 methods
were compared in mice to determine which
represented the most quantitative delivery A major concern in dosing rodents with test
system, For dietary incorporation, a high. compounds isthe accurate quantification of oral
moisture bolus form of NIH.31 rodent meal was delivery. The water bottle delivery method
developed using hydroxypropyl methyleellulose continuoufly supplies water-soluble compounds
u an atttoclave-stable binding agent. A high- over a period of time. Problems limiting the
moisture bolus was selected to increase the usefulness of water bottles include test com-
acceptability of the dosed diet and to promote pound palatability and solubility, spillage (Lane
quantitative consumption through reduced et ai., 1984), soundness of the bottle stopper,
wastage, The test compound u_od wa_ D.xylose, leaching of compounds from the stopper
a pcnlo_,e sugar that may be quantitatively (Kennedy &Beal, 1988), and individual variations
detected, coloflntetricaliy, In urine following in water demand (Weisburger & Weisburger,
oral dosing. Six male and 6 female B6DgFI 1967). Gawge delivers a known quantity of test
mice were placed tn metabolism cages _nd compound in asingle dose. The disadvantages
dosed with a known quantity of D._ylo_e by of gavag¢ dosln$ include oesophageal orstomacl_
e_teh of the 3 ntethods, Urine was collected damage, intubation of the lungs, and large,
before and after each method of administration potentially fatal spikes intest compound plasma
and analysed for total D.xylose; tile per cent concentration (Welsburger &Welsburger, 1967;
recovers was b_ed upon the amount of D.xyio_ Lindamood et el,, 1988). The gavage dosing of
consumed, Quantitative consumption was animals isalso very labour.lntensive.
apparently greatest for water bottle do_ing with High-moisture diets contain a binder that
an average recovery of 56' 0% ofthe original D- combines water and the diet meal into a semi-
xyio_ dose, High.moisture bolus Incorporation solid mixture. This mixture isa useful method
ranked _ond with _0,0_/0 D.xylo_e recovery, for presenting dusty, volatile, or toxic test
and garage was third writ 41'0% D-xylose compounds to anlmal_ with minimal spillage and
recovery. wastage, thus reduclng the risk of exposure to
toxic test compounds to the technical staff (Lane
et al,, 1984; Clapp &Bradbrook, 1982), If the
Correspondence to: Dr SM Lewit, test compound is pro-mixed into a soluble diet
ingredienstu,chas lipophili¢ compounds mixed
R_celwd )8 December lgPl; accaptca POctober lPJP2
l
'D,
",4 "
._-2?-19_.3 15:_9___ ....... BIO-3 407 853-4220 CCAFS P,02
Dosing methodl comparieon 165
with the fat component, water-soluble or in a pre-fceding bolus, and to compare it with
.Insolublecompounds canbeIncorporatedInto garage and water bottle metltods for efficiency
the diet (We_burgcr &Weisburgor, 19M), Studies inoral delivery of a known quantity of D-xylose
usinghigh-moisture, semt-purlfied dietscontaining to B6D2F1 mice. The bolus was formulated with
agar as a binder have been successfully used to acommercialfoodbinder andtestedforaccept.
feed mice (Lang et al,, 1984), rats (Ciapp & ability by male and female B6D2FI mice (Mus
Bradbrook, 1982),midguine_plgs 0Navia&Lopes, musculus), D-Xylosc, a pentose sugar, was
1973),Recentstudiesshow agarinducesnon- chosen as a test compound based on its water
pathologicaplh,ysiologiccahlmigcsincaecumand solubilityr,apidurinaryclearance(Craig&
colon cell growth; however, itmay promote the Atkinson, 1988), lack of toxic effects, ease of
carcinogcniqjotfyccrtalcnompounds (Shiau& detection (Eberts el al., 1979), and because its
Wang, 1988). Compared withgarage, dosed diets absorption from the intestinal tract is pro-
have been used to deliver higher levels of an portiontd to the dose given (Stradley et al._ 1986).
unpalatable and highly toxic compound with a
reducedmortality rate(Lindamood elal,, 1988). Materials and methods
The objectives of this study were to develop High-moisture diet formulation
ahigh-moistur¢, natural-ingredient dietary form Eight food-processing companies supplied 20
usefulforefficientplryovidingtestcompounds commercial food binders forthis study, The classes
Table l, Commercial food bi_der_ evaluated
Commercial
Producl name Binding o#ent Selecled binder aourc¢ _
H-SO Cassava stacch +
IF-I31 Cassava starch -
National 78.1272 Cassava Starch
Redisol 412 Cassava starch -
Redisol 24B PotalO starch -
Thtn-n-Thik 99 Corn starch
Sta-Ml=t 36_ Corn slarch -
eta-Mill 4.$4 Corn starch -
MethoceI (MC) Meth_.lc¢llulose -
Meth_el (HPMC} 14ydroxypropyl methylcellulose +
CMC R-7_-H4 Sodium carboxymethylc¢llulose ÷
CMC R-95-H4 Sodium _arbox),methylcellulose -
Avice[ PH-101 Mlcrocrystalline cellulose -
Avicel RC-591F Microcrystalline cellulose -
RhodtEel Xanthan gum -
Polydeatrose Dextrose polymer -
1701Dextrin Corn dextrin -
Lo-Dex 10 Corn molto-dextrin -
1620Oextrlfi Corn dextrin
"1710 Dextrin Corn dextrin -
'Binders were selcc|¢d based upon selection criterion pr¢-©stablished for this research; selection is not a test of producl
endorsement,
"=National Starch and Chemical Corp., Brfdgcwatcr, NJ.
CA,E, Sial=), Manutat:lurln=, Decatur, IL,
_Dow Chemical, Midland, MI. !
'Louisiana Chemical Polymers, Baton Rou_e, LA.
=FMC Corporation, Philadelphia, PA.
*Rh6ne-Poulenc, Monmouth Junction, NJ.
_Pftzer Chemical Division, New York, NY.
'Am¢rlcan Mails.Products, Hammond, IN,
+Acceptable or - non-acc0ptable bytest criteria,
-£7-1993 15:39 BIO-3 4@7 853-4££@ CCAF8 P.O3
166
Zimmer, Lewis &M0yer
of binders supplied were as follows: chemically- methylcellulose (Louisiana Chemical Polymers,
modified and unmodified tapioca starches, Baton Rouge, LA) and H-50 modified cassava
modified _mdunmodified corn starches, modified starch (National Starch &Chemical Corp., Bridge
potatostarch,hydroxypropylmethylcellulose,water, NJ), The 3 mixtures used in the animal
methylceilulose, microerystaillne cellulose, sodium acceptability evaluation were' (1)a 1:3:5 ratio of
carboxymethylcellulos¢, xanthan gum, poly- binder:diet :water using cassava starch; (2) a
dextros0, malto-dextrln, and modified dextrins 1:5:7mixture using hydroxypropyl methylcdlu..
(Table 1),
lose; and (3) a 1:5:7 mixture using carboxy- '
Autoclave stability of the binders and their methylcellulose (Table 1),
usefulness In high-moisture diet applications
were determined by practical tests. Various Evaluation of animal acceptability
combinations and varying proportions of binder, An acceptability trial was conducted to test the
diet and water were autoclaved to determine boluses using 12 5-month-old B6D2FI mice of
physical stability and qualitative properties of the conventional microbiological status. Animals
diet bolus formed. Binders were first autoclaved wereallocated to one of 3groups, 2males and 2
indlvidually to determine tl_eirability to retain females ineach group. All animals were acquired
pre-autoclaving form and consistency. Binders from the National Center for Toxicological
exhibiting physical changes or deterioration after Research, Jefferson, AR, Breeding Facility. The
autoclaving were eliminated from further con- boluses contained I g of ground (20 mesh)
slderation (Table 1). NIH-31 diet hand-mixed with binder and water
Binders exhibiting stable properties were then in the ratios determined in the previous stage of
autoclaved with ground (20 mesh) NIH.31 stan- testing, Environmental conditions of the animal
dard rodent dlet (Purina Mills Inc., Richmond, rooms were 23_+1.5 '_Ctemperature, 50± 9%
IN) In dry premixes containing 1, 5, or 10070 relative humidity, automatic 12:12 ligl_t:dark
binder, by weight. The NIH-31 diet was chosen as cycle (lights on at 060010, and HEPA filtered
itremains physicallystable when autoclaved and is air with 10-15 exchanges/h, The animals were
a natural-ingredient, completely balanced diet, individually housed; food and water was available
Animals were maintained on pelleted NIH-31 diet ad ltbitum before and after the testing. The mlce
when not on test. Dry premixes that did not were fasted for 24h immediately prior to testing
undergo physical changes or deterioration were to promote complete and rapid consumption Of
then hydrated with 5ml aliquots of water, added the bolus. Polycarbonate shoebox cages were
Incrementally, and hand.mixed until abolus was modified forthetest by gluing a small, pre-
formed, If a cohesive bolus dld not form when weighed beaker containing the bolus to tile cage
a total of 1$ml had been added to the dry wall one-inch from the floor to prevent con.
premix, no further water was added. tamination of the bolus with faeces or urine,
Mixtures of binder, diet, and water were then Bedding was removed from the cages during the
formulated for autoclaving and evaluation. I hof exposure to the bolus inorder toaccurately
Mixtures contained 1, 5. or 10tr/0of binder by observe the animals' acceptance, The animals
weight. Water, as 33, 50, or 60°70of the total were continuously observed todetermine quantity
weight, was added to the dry ingredients prior to consumed and wastage or spillage of the holus.
autoclaving, resulting in 9 samples ttst¢d per
binder. Binders to be tested for animal accept- Evaluation of dosing efficiency
ability represented 3 chemical classes and were Twelve male.and 12female 5-month-old B6D2FI
superior in autoclave stability and cohesive mice were used in this section of the study.
properties. The 3binders selected were Methocel® Environmental conditions were the same as in
hydroxypropyl methylcellulose (Dew ChemicaI the evaluation of animal acceptability, The
USA, Midland, MI), CMC R-75-H4 carboxy- animals were allocated to 2 replicate groups of
'"',,
,-27-1993 15:40 BIG-3 487 853-4228 CCAFS P,84
j,2/'
Dosing methods comparison
/¢ 167
6 males and 6 females. While one group was returned after the treatment. NIH-31 pelleted
being tested; the other group was housed in diet was available ad iibitum throughout this
polycarbonate shoebox cages with ad llbitum treatment.
food and water provided. During collection The garage dose was 0.5 ml of a 5'08 D-
intervals, animals were housed inpolycarbonate xylose/50ml water solution (0.05 g D-xylose/
metabolism cages (Maryland Plastics Inc., dose) administered via syringe and blunt-ended
Federalsburg, MD) and acclimatized for 3 days garage needle. Water and NIH-31 pelleted diet
prior to the trial.
were available ad iibitum throughout this
The test was conducted in consecutive treat- treatment.
ment order with each animal receiving all
treatments with intervening intervals inwhich the Sample analysis
effects of the previous treatment were determined
The urine samples were analysed using a
by urine assay as described below. Testing o(the modified colorlmetric method of Eberts (Eberts
high-moisture bolus, water bottles, and gavage etal., 1979) in which benzoic acid was omitted
was performed in consecutive 4-day periods, from the assay. Five standards containing 0,5,
Urine was collected on the first day (Day 0) of 1.0, 2'5, 5'0 and 10,0 mmol/l of D-xylose were
each period to establish baseline D-xylose prepared to establish daily calibration curves.
excrctlon valuesm_dto confirm thatD-xylose ex- Consistent linear calibration curves were estab.
cretion had returned to baseline values before lished usingstandards prepared bythemodified
beginning another treatment. The dose was procedure, Samples were prepared in 35ml
provided after baseline urine samples were screw-top tubes with Teflon©-lined caps and
collected and analysed. Urine was collected incubated in a boiling water bath. When the
for 3 consecutive days following dosing to O-xylose content of a samplewas below detection
ensure complete recovery of the D-xylose. To limits, the assay was repeated using 2or 4 times
prevent bacterial growth in the urine samples, the urine concentration to achieve a measurable
0.2ml of a 100/0thlmerosal solution (Aldrich
D-nylose concentration. Results were adjusted
Chemical Co., Milwaukee, WI) was added to
for the increased urine concentrations. All samples
each sample prior to analysis (J Knowles, were runintriplicate and averaged, samples were
personal communication). reanalysed if the variance exceeded 10%.
The high-moisture diet bolus, 63°70drymatter,
contained !.0g ground (20 mesh) NIH-31 rodent Statistical analysis
meal, O.1g Methocel@hydroxypropyl methyl- The main effects of sex and treatment upon D-
cellulose as the binder, and 1.6 ml of a 6.25 g xylos¢ recovery between replications were the
D-xylose/100 ml water solution to provide 0.1 g factors considered. Statistical analyses were
D-xylose/dose. The boluses were mixed by hand performed byusing analysis of variance obtained
and provided to the mice for 24h, Consumption from the General Linear Models procedure of
wasmeasured as change in weight of the beaker
Statistical Analysis Systems (SAS, 1982) using
and contents. No spillage or wastage was noted. least squares calculation of treatment means and
Water was available ad llbitum throughout this F-protected comparisons. Tukey's test was also
treatment. TheNIH-31 pelleted diet wasreturned employed to verily multiple means comparisons.
following bolusconsumption.
The water bottle dose consisted of 1.0 g D- Results
gylose in 100m! water placed inthe water bottles Evaluation of animal acceptability
for a 24-h period. Intake was represented as a
The criteria for evaluating the boluses were: (1)
change in weight of the water bottle and consumption bythemice and, (2)cohesive integrity
contents. The concentration of D-xylose con- of the bolus while the mice were feeding. The
sumed was then calculated. Fresh water was bolus containing cassava starcll was consumed
°,,°
•_-27-1993 i5:4i BIO-3 4Q? 853-4_0 CCAFS
P.@5
161
Zimmer, Lewis&Moy_
by halfofthemiceand showed highcohesive
Table 3, Percent daily x¥loz¢recovery from hlgh.mohJlure
integrityT.he boluscontainingthecarboxy.
hoh4s, garage tnd water bottle treatments among S-month,
methylcellulosweas consumed by allof the old B6DZFI mice
mice butshowed poorcohesiveintegritTyh.e
Dally per ¢¢nt recovery
bolus containing hydroxypropyl methylcellulose Treatment 1 2 3
was consumed by 3 of the 4 mice and showed
Hlgh-moi_tu|'©
high cohesive integrity. Based on these results,
bolus 51,8' 2.2 O.16
hydroxypropylmethylcellulowsaesselectefdor Gavase 44'9 _ 1,3 0,97 J
use in the bolus preparation for further testing. Water bottle 57.9 _ J,9 -0,|8
SI_M 2,09 i .04 0,634
Hydroxypropyl methylceilulos¢ has beenproven Male High-moisture
safeforlong-andshort-termusewlthrodents bolus 40'2 "_ 6.1 -0,30
",t'
(WHO, 1974), Oavalte 32. S# 2, I 0,87
Water hotfie 44.3" 6.3 I,S9
$EM 4,00 _.52 0,683
EvoCationofdosingefficiency
Therecoveryefficienocfyeachdosingtreatment "bMeans tn a column with different superscripts differ
(P,_O,05),
was calculateads per centof the original D-
xylose dose which was recovered in the urine
methods tested. Recovery among females for the
samples. The average D-xylose dose received by
high-moisture bolus,gavage and water bottle treat.
,I
each animal was 0.091 +0.010g for animals
4 manta differed P< 0, I0,P< 0. Ofand P< 0,05,
consuming the hlgh-molsture bolus, 0.042±
respectively, as comparedwith malerecoveries.
0,012 g for animals receiving water bottles, and
Among females, significantly higher (P< 0,05)
0.05 g for garaged animals (no variance). Two
total D-xylose recoveries were found for tile
male mice died during the study. During the first
water bottle and high-moisture bolus methods,
replication, one mouse died from unknown
59.6 and 54.1070, respectively,ascomparedwith
, !
causes after the final water bottle treatment
garage, 47. I_/o.Among males, slgz_flcantly higher
period. During the second replication, one mouse
(P< 0.05) total D-xylose recoveries were found
died from accidental lntubatton of thelungs
for the water bottle and high-molsture bolus
during the garage procedure. Statistically, the
methods, 52.5 and _,0%, respectively, as com-
losses only reduced the observations for the
paredwith savage, 35.5070. The trends for both
Savagecomparisonwheren- IIforthe males,
sexesshowed a similar hierarchy dueto treatment
As therewasno statisticdiafflerence due to
(water bottles >high-moisture bolus > gavage).
replicatiorne,sultwserepooledforcomparison
The D-xylose recoveries for the first day (Table
of per cent D-xylose recovery by sex and
3) following each treatment showed significantly
treatment (Table 2). Females had significantly
(P<0.05) greater recoveries for the high.
higher D-xylose recoveries than males for all
moisture bolusand water bottle treatments as
compared to garage dose recovery,• 51.8 and
Table 2, Pc'rcent recove_ of total xyiott dou from hlllh- 57'90"/0vs44.9_0, respectively. For males, water
"molwture holtis, llavtSe and water botlle trealmenlt among bottle and high-moisture bolus treated animals
S-month-old BeD2FI mice
demonstrated significantly (P<0.05) more
Treatment,_tRecovery efficient first day D-xylose recoveries than
Ifl_h,
moisfure Wafer garage, 44,3 and 40.2O7ovs32.5We, respectively.
$ex bolus Oavo&e bottle $1_M Essentially, all D.xylose recovery was complete
by 2 days post-treatment.
i.
Femtle 54. I_ 47-Is 59"6' 2"30
Male 46,0 _ 35'5 s $2,5' 2"87
Discussion
*'SM_ans In a row with different sttpef_crlpts differ The 3 methods for oral administration of test
(P_O,05),
compound used in tiffs study have various
• -.i.... I.L-:.:- .... .........
. • . s, ..
",,
•'..',. , '. . ,'
.,-27-1993 1_:42 . BIO-3 487 853-4220 CCAFS P. 08
%/
/
169
Dosingmethodscomparhon
advantages and disadvantages that determine sugars may be converted, in part, to glucose or
their suitability for a study. When the com- may enter glycogencsis via the pentose phosphate
parative efficiency of gavage, water bottles, and pathway. This latter conversion had been re-
a novel high-moisture bolus in delivering a ported to occur in the Intact mouse tHiner,
known quantity of D-xylose to male and female 1957).
B6D2FI mice was analysed, water bottle A significant sex difference was seen in all
delivery demonstrated the highest D-xylose treatments inwhich D.xylose recovery among fe-
dosing efficiency and recovery of the 3methods. males was consistently higher than among males.
However, the efficiency of the high-moisture A study investigating sex differences in human
bolusdelivery was not statistically different from D-xylose excretion found a tendency among
that of the water bottles. fetnales to excrete D-xylose more efficiently than
For each dosing method tested, there are males after an intravenous dose (Kendall &
associated disadvantages. For the high-moisture Nutter, 1970), Similar research has not been
bolus, the bolus must be given separately from previously conducted for mice.
normal feeding regimens, requires added labour, As garage dosing is widely used as a method
and a pre-fast interval to ensure complete of chemical administration, the results reported
consumption. For the water bottles, dose loss here are of particular interest. The comparison
due to spillage isa commonly cited disadvantage of other compounds among the 3 dosing
(Long etal., 1984). Inthisstudy, one animal died methods is warranted to provide more in-
after accidental intubatlon of the lungs following formation about the recovery effieiencies of the
garage, methods studied.
It isnot known whether the binder used in the Xylose isthe chief pentose that is actively ab-
high-moisture bolus had any inhibitory effect on sorbed from the gut (Roehrig, 1984). Absorption
D-xylose absorption; however, gel.formlng gums, of the total D-xylose dose may have been
including carboxymethylcellulose, have been inhibited by the saturation of the active transport
shown to reduce D-glucose transport in the rat system. Although D.xylose was selected for its
jejunum (Johnson &Gee, 1980. If D.xylose and previously reported stability, perhaps thispemose
D.glucose share a common transport pathway, sugar is metabolized as previously suggested
as suggested by some researchers (Ohkohchi & (Hiatt, 1957).
Himuki, 1984), the inhibition el"transport bythe The summarized data show that both water
hydroxypropyl methylcellulose gel used could bottles and the high.moisture bolus are com-
have reduced the efficiency of D-xylose ab- parable in delivering a known quantity of D-
sorption from the hlgh-molsture bolus. Based on xylose to mice. Water bottles showed a con-
D-xylose absorption studies using rats, the minor sistently higher, but not significantlydifferent,
differences in concentrations of D-xylose efficiency than the high-moisture bolus, but the
provided by the garage and water bottle method islimited to water.soluble compounds.
solutions (0.05 and 0.042 g, respectively) should The high.moisture bolus is more versatile with
not have changed the proportion of D.xylose regard to compound solubility.
absorbed (Stradley etal., 1986).
The greatest recovery for D-xylose observed
inthis study wasan average 56°/0following water Acknowledgments
bottle administration (Table 2). Earlier The authors wouldlike to express their apprecia-
researchers (Segal & Foley, 1959) report that tion to the Office of Research Services, especially
labelled D-xylose infusion in man resulted in an the Divisions of Microbiology and Chemistry, at
average 44% urinary recovery of the total dose. the National Center for Toxicological Research
These researchers proposed that the pentose for support in this research.
L
'ol. w
170 Zimmer, Lewis & Moyei"
Referenges
Cl_ppMJL &Bradbrook C(1982)Growth andlongevityof N,_vlaJM &Lopez H (1973)A purifiedB¢Idietforguinea
rats fed an ugar-bound diet, £aboratory Anhnals 16, pll_S,Laboratory Animal Science 23, 11i-114
|38-142 Ohkohchl N & Hlmukal M (191i4) Speciesdifference in
CraillRM &AtklnsonAJ (198R)D.xyloletestingA:review. mechanisms of D-xylo_ Absorptionbythesmall Intestine.
Gaslroetllerolox._ 95, 223-23 l Japanese Journal of Ph.vsiology 34, 669-699
Ebem TJ, SampleRI-IB,{3licMkR, elat,(1979)A dmpilfied, Roehrlg KL (1984) Digestion of simple carbohydre_te_, In
calorimetric microme[hod for xylosein serumor urine, Carbohydrnle Btoohemt_lr.V and Metabofism. Westport
with ph[orogiuclnol. Clinical Chemistry 25, 1440-1443 CN: AVI Publishing Company, lnc,
Hiatt HH (1957) Olycogen formation via the pentose Saga[ S &Foley JB(|959) The metabolic fateof Cj_labeled
phosphatepathwayinmiceIn vlvo. Journalq[ Blologleal pentoses in man. American Journal of Clinical Investiga.
Chemistry 224, 851-8:59 lion 311,40'/-413
JohnsonIT&Gee JM (1981)Effectofgel-formlnggums on Shiau SY & Wang HJ (i985) Effects of dietary agar on rat
theinteitJnal unstirredlayerandsugartrane_portin Vitro, colonic and cecummucosal growth. Nutrition Reporls
Or# ;_2,398_403 lnlRrnallotwl 38, 147-15_1
KendallMJ & NuttcrS(1970) The influenceofsex, body Statistical Analysis Systems,The GLM Procedure (!982) In:
weight, Andrenal function on the XyIo_n lest, GUt 11, SAS User's Guide." Stall, tic#, Version 5, Cary, NC:
i020_1023 Stiltlslica[Analysis SystemsInstitute, [he,, 433-_06
Kennedy BW &.Beal TS (1988) Mineral contribution to S[radley R,Farmer-I_Jley P, Pa_uini N,eld. 0986) Gastric
drinking waterfrom rubberstoppers, Laboratory Animal absorptionofD-xylme inthe rat: its influence on_heD-
Science311,49"/-498 xylose absorption test, Journol of Laborator.v and Clinical
Knowles J, National C_nl¢r for Toxicological Research, Medicine 107, 10-14
MicrobiologyDivision,]'arsenaclommuuieatlon Welsburger JH & Weisburger I_K(1967)T¢_ts forchemica[
Long JA, [.o,;C:M;g& Whh¢ WJ (19;14U)icat'_I_;_r-based carcinogens, ln: Methods in c_ncer reason'h° Vat 1 (ed.
diet to fulfiJ the food and water requirements of mice, H lluseh)pp. 307-398. New York, NY; Academic Pres_
Laboralory Animals t8, 40-41 WorldHealth Organization (1974) Toxicolo_ti_t eval_llO_
Lindgmood C,Lomb JC,llrlslDoWl, el aL (1988)Studies of some food additives lnt,ludln_ antlc'aking agent#,
ontheshort-ferat0lx)clty ofthcophylllnin¢r_l_andml_e. antimlcrobtats, antioxidanls, emul_'lflvrs arid lhtckentag
Fundamental and Applied ToxicologY 10, 4'/7-489 agents. WHO Reporl Wl WI4H rio 5m297-]15
L
(-),:/i2T7 '9:3 15:50 PAGE. 007P
