Is Fructose Sweeter Than Glucose for Rats?

Physiology& Behavior,Vol. 60, No. 5, pp. 1299-1306,1996Copyright0 1996ElsevierScienceInc.Printedin the USA.All rightsreserved

0031-9384/96$15.00+ .00

PII S0031-9384(96)00259-4

Is Fructose Sweeter Than Glucose for Rats?

ISRAEL RAMIREZ

Monell Chemical Senses Center, Philadelphia, PA I9104-33O8 USA

Received 17 November 1995

RAMIREZ, I. Is fructose sweeter than glucose for rats? PHYSIOL BEHAV 60(5) 1299–1306, 1996.—Because it is generallythought that the intensity of the taste of fructose is greater tharr that of glucose for rats, it seemed surprising when sham-fed ratsdrank substantially less of a mixture of 6% fmctose plus saccharin than of a mixture of 6% glucose plus saccharin. At least 3different factors contribute to this effect. First, the taste of fmctose is Iess attractive to rats thasr is the taste of ghrcose; sham-fedrats strongly preferred glucose over fructose (no saccharin was used in this experiment). The second factor is experience. Ratshaving substantial previous experience with glucose, but not with fructose, consistently preferred glucose over fructose. Conversely,rats having substantial previous experience with fructose, but not with glucose, initially showed no consistent preference butsubsequently tended to prefer glucose. The third factor is an interaction between saccharin and the type of sugar. Rats given onlyone solution at a time drink approximately as much fructose as glucose when the solutions contain no saccharin. The addition of0.25% saccharin to 6% glucose stimulated intake, whereas the addition of the same amount of saccharin to 6% fructose did notstimulate intake. As a result, rats ingested substantially more of a mixture of 0.2570 saccharin plus 670 glucose than they did of acomparable mixture of saccharin and fructose, even though rats ingest similar amounts of fructose and glucose without saccharinin single-bottle tests. Because the differential effect of saccharin on intake appeared within 2 h in naive rats, and did not greatlychange over a 3-day period, it is probably not attributable to conditioning. These results suggest that these sugars have qualitativelydifferent tastes. Copyright O 1996 Elsevier Science Inc.

Sweet sugars Preference Acceptance Sham feeding

IT HAS long been known that the ability of sugar solutions tostimulate ingestion is at least partly attributable to their taste( 14,16,38). Recent studies indicate that the postingestive effectsof sugars also contribute to their ability to stimulate intake( 11,32,34,35,44). However, the postingestive effects of fructosehave been claimed to have a weaker motivational effect than dothe postingestive effects of glucose ( 1,2,39,40), but see (48).To account for the observation that fructose often stimulates in-gestion more than does glucose (5,8), it therefore seems neces-sary to assume that fructose is a more potent stimulator of gus-tatory responses than is glucose, something that has beenhypothesized on the basis of other evidence (5,17,21,29,47).These concepts neatly account for much of the data in the liter-ature.

The present series of experiments was initiated on the basisof the assumptions described above. The original intent waspartly to make use of fructose’s greater taste intensity, and partlyto obtain pilot data for studies of the effects of intragastric in-fusions on saccharin intake. These experiments produced resultsthat seemed difficult to reconcile with previous concepts aboutthe factors governing intake of fructose and glucose.

GENERAL METHODS

Animals

Female CD rats (Crl:CD@BR, Charles River Breeding Labs.,Wilmington, MA) arrived in the laboratory when they were about8 weeks old. They were housed individually in hanging stainless-steel cages, maintained on a 12/12 h light/dark cycle at 21–24”C, and given Purina@ Laboratory Chow (#5001) ad lib, ex-

cept as indicated below. When not given test fluids, the rats weregiven tap water in sipper bottles to drink ad lib; they always hadsome fluid to drink. The rats were maintained in the laboratoryfor 1 week after arrival to permit them to acclimate to their sur-roundings before testing began.

GeneralProcedure

Food was in steel hoppers in the rear of the cages. Fluids wereoffered in 120–240-ml glass jars placed inside the front of thecages. All mixtures were formulated on a weight basis becauseintake was measured by weighing. Solutions were made freshdaily with deionized water.

Statistics

All group data are given as means 3 SE. Statistical analyseswere based on simple intake. When a choice was available, intakewas analyzed by repeated measures analysis of variance. Anal-yses based on preference scores gave generally similar results,but are not reported here. Between- and within-group compari-sons were done by analysis of variance and t-tests. Probabilityvalues were computed by a nonparametric method, variouslyknown as permutation, randomization, or resampling ( 13,24).

EXPERIMENT 1

The original purpose of this study was to examine oral satietyin sham-feeding rats. It was hypothesized that it might be pos-sible to stimulate intake of a dilute sugar solution by adding sac-charin (43 ) and/or substituting a sugar thought to have a greater

1299●

1300 RAMIREZ

taste intensity, fructose. The test sessions were relatively long ( 1 periment 2 began. All rats were put in the test chambers for 4 h,h) to allow the rats ample time to satiate. had their cannulas flushed, returned to the test cage for 1 h, and

were then offered a choice of 6~0 glucose vs. 6?I0fructose (noMETHOD saccharin in either solution) for 1 h. AH cannulas were open

Stainless-steel gastric cannulas were implanted into 12 ratsusing procedures similar to those described by others (6,22),except that the cannula exited through the original incision andnylon suture was used to minimize infection. The rats were al-lowed to recover for at least 1 week after surgery before testingbegan.

On each of 3 consecutive days, the rats were put in Plexiglastest chambers with no food or water for 4 h, allowed to drink 690glucose for 2 h and then returned to their home cages where foodand water were available ad lib. Fluids were presented at thebeginning of the dark part of the light/dark cycle, shortly afterthe lights went out. The cannulas were closed during these train-ing trials. Three of the rats drank very little glucose during initialtests and were therefore allowed to drink the solution in theirhome cages. By the third training day, all rats drank glucoseduring the first h.

Following a 2-day rest period, the rats were given sham-feed-ing trials on 3 consecutive days. On each day, the rats were putin the test chambers with no food or water for 4 h, had theirstomachs flushed with warm tap water, returned to the test cagefor 1 h, given a sugar solution to drink for 1 h (at the beginningof the dark part of the light/dark cycle), and then returned totheir home cages. The cannulas were open during these trials. Onthe first 2 days, the rats were offered 6% glucose to drink. Onthe third day, the rats were given either 6% glucose or 6% frnc-tose; both solutions contained 0.25Y0saccharin in addition to thesugar.

RESULTSAND DISCUSSION

On each of the first 2 sham-feeding tests, the rats drank 18t 2 g. Contrary to expectation, rats given fructose + saccharindrank a mere 2 * 0.6 g, whereas the rats given glucose + sac-charin drank 19 ~ 4 g (p = 0.002). Thus, intake of glucose +saccharin was approximately equal to intake of glucose withoutsaccharin on the previous day. The absence of overingestion ofglucose + saccharin and the extremely low intake of fructose+ saccharin seemed so contrary to seemingly well-establishedideas that the experiment was immediately terminated. Theamount of fluid that drained from the cannulas was similar tothe amount ingested, except on the first day when 1 rat had aclogged cannula.

EXPERIMENT 2

Experiment 2 examined preference for glucose vs. fructose ina group of rats that had previously been habituated to sham feed-ing. The purpose was to determine whether rats would ingestmore fructose or glucose. Because the surprising outcome of thefirst experiment might have been attributable to saccharin, noartificial sweeteners were used.

METHOD

Experiment 2 employed another set of 12 rats. Surgical tech-nique and test protocol were similar to that of Experiment 1.However, the rats participated in another experiment before theyparticipated in the tests described here. In the first experiment,the rats were given a series of 7 tests with 6Y0glucose, in whichhalf of the rats always had their cannulas open and the other halfalways had their cannulas closed. All rats were then offered 670glucose for 24 h with their cannulas closed. Two days later, Ex-

during this test. This procedure was repeated on the next day,”butwith the left right positions of fructose and glucose reversed.

RESULTS

The results of the initial portion of the experiment, in whichthe rats were only offered glucose, are not described here becausethey are not directly relevant to this report. The total amount offluid drained from the cannulas during the 2 preference tests was25 t 2 and 23 3 2 g, which is slightly less than the amountingested, 25 t 2 and 24 * 3 g. Rats ingested substantially moreglucose than fructose (p <0.001 on both days, see Fig. 1). Forthe average of both days, this effect was greater in rats that hadpreviously ingested glucose with their cannulas plugged than forrats that had previously ingested glucose with their cannulas open(p = 0.04, but p = 0.07 on the first day, and p = 0.12 on thesecond day). The 2 groups did not differ with regard to total fluidintake (all p = 0.13).

DISCUSSION

Although some sugar solution was probably absorbed (42),this should not greatly alter the interpretation of the experimentbecause the amount of sugar absorbed must have been small. Theapparent difference between the 2 groups might be attributableto sham feeding having an aversive effect, or to reaf feedinghaving a rewarding effect (45,50). For the purpose at hand, how-ever, the critical point is that both groups preferred glucose overfructose.

EXPERIMENT 3

If one assumes that rats prefer the more intensely flavoredsugar solution, the results of Experiments 1 and 2 would suggestthat the taste of glucose is more intense than that of fructose.Because the postingestive effects of fructose are thought to be

_ Fructose Glucose

30[ T25 -

~L> 20 -

$ ,5 -mc—u 10 -?ii

5 -

Plugged Open Plugged Open

1st Test 2nd Test

FIG. 1.Mean f SEMshamintakein Experiment2. The rats were givena choice of 690 fructosevs. 6?i0fmctose for 1 h on 2 consecutivedays.Thepluggedandopengroupswerepreviouslygivensham-feedingtrain-ing with their cannulaspluggedor open.All rats had opencannrdasforthe tests depictedin this figure.

FRUCTOSE AND GLUCOSE

~ Fructose

100Fa [

: 4oL---&

-e-- Glucose

1301

for 2 days, during which they had plain water to drink, and thenwere given the same solutions in descending sequence. Each so-lution was available for 24 h at a time.

RESULTS

A’There were no significant differences between the groups with

1 regmdtoint&eofsugmsolutions,evenwhenline~kendanal-yses were applied (all p >0.2, see Fig. 3).

EXPERIMENT 5

0/0 Sugar Concentrateion

FIG. 2. Mean t SEM 24-h intake of sugar solutions in Experiment 3.

less rewarding than those of glucose ( 1,2,39,40), a real-feedingtest would be expected to demonstrate even greater differencesbetween glucose and fructose. Experiments 3 and 4 attempted toverify this prediction.

METHOD

Four rats were offered an ascending series of fructose solu-tions over 4 consecutive days (O%, 2, 4%, and 6Yo). Another 4rats were offered the same ascending series of glucose solutions.The rats were rested for 3 days and then retested in the samesequence, except that the rats that were previously given fructosewere given glucose, and vice versa. Only one solution was avail-able per rat on any day and each solution was available for 24 hat a time.

RESULTS

As might be expected, fluid intake increased with sugar con-centration (see Fig. 2). Rats ingested similar amounts of glucoseand fructose solutions. There was not even a reliable tendencyfor the difference between fructose and glucose to increase withconcentration (linear trend ignoring order of sugar presentationp = 0.07, linear trend taking order of sugar presentation intoaccount p = 0.09). Other tests comparing fructose and glucosefailed to approach significance (p >0.2, see Fig. 2). Rats givenfructose first did not differ from those given glucose first (all p> 0.2).

EXPERIMENT 4

It is possible that the failure of the previous experiment toshow significant differences between the two sugars might havebeen due to the use of a within-subjects design, or to the use ofoverly low sugar concentrations. Experiment 4 employed a be-tween-subjects design and included sugar concentrations as highas 12%.

METHOD

A new group of 14 rats was divided into 2 groups that weregiven either fructose or glucose. During the first week, the ratswere offered the following sequence of sugar solutions over suc-cessive days: 3~o, 6Y0,9%, and 12910.The rats were then rested

The absence of a large difference between fructose and glu-cose in Experiments 3 and 4 seemed inconsistent with the largedifferences observed in Experiments 1 and 2. Because the testsin Experiments 1 and 2 were much shorter than those in Exper-iments 3 and 4, it was hypothesized that test duration was thecrucial factor.

METHOD

The rats from Experiment 4 were given water only for 4 daysand then offered 6’%0fructose or glucose for 30 rnin (each rat wasgiven the same sugar that it had received on previous tests). Thiswas repeated on the following 2 days, except that the test durationwas shortened to 10 min. On the fourth day, the rats were givena choice of 6% fructose or 6% glucose for 10 min. Tests wereconducted at the beginning of the dark part of the light/darkcycle, approximately 15 min after the lights went out.

RESULTS

There were no significant differences in the 10-min single-bottle tests (p ~ 0.08, see Fig. 4). During the choice test, ratsdrank more glucose than fructose overall (p = 0.02). When thegroups were analyzed separately, this effect was significant forthe glucose group (p = 0.03), but not for the fructose group (p= 0.11, see Fig. 4). However, the fructose group drank morefluid (glucose plus fructose) than did the glucose group duringthe choice test (p = 0.04).

DISCUSSION

This experiment confirms the finding from Experiment 2 thatrats having prior experience with glucose prefer glucose over

~ Fructose --0-- Glucose

100[

80 -

60 -

40t

t20’

o 3 6 9 12

0/0 Sugar Concentration

FIG. 3. Mean f SEM 24-h intake of sugar solutions in Experiment 4.

1302 RAMIREZ

u.—3ii

= Fructose

‘[ I

6 -

3 -

o—

T

Glucose

T

Test 1 Test 2 Test 3 F–Group G–Group

Single–Bottle Tests Preference Test

FIG. 4. Mean * SEM 10-rnirt intake of sugar solutions in Experiment 5.Tests 1 to 3 were single-bottle tests. The bars on the right depict a singlepreference test.

fructose. It is also consistent with Experiments 3 and 4 in thatrats drank similar amounts of fructose and glucose in single-bottle tests.

EXPERIMENT 6

Experiment 1 demonstrated a large difference in intake offructose and glucose that has not been confirmed in subsequentsingle-bottle tests. It seemed possible that the presence of sac-charin was crucial. Another possibility was that frequentlychanging the sugar concentration in experiments 3 and 4 blockedthe effect. Experiment 6 assessed these possibilities.

METHOD

Twelve new rats were divided into 2 groups of 6. One groupwas given fructose and the other group was given glucose todrink as their sole fluid source on days 1, 2, 3, 7, 9, 10, 14, 15,16, and 17. Plain water was available on other days. The sugarconcentration was 6?t0on days 1–3 and 7– 10, and 3’%0on days14–17. On days 9 and 16, the sugar solutions contained 0.25%sodium saccharin.

RESULTS

On every day except day 9, the 2 groups drank similaramounts of fluid (all p > 0.2, see Fig. 5). On day 9, the groupgiven 6% fructose plus saccharin drank substantially less fluidthan did the group given 6% glucose plus saccharin (p = 0.03,see Fig. 5). Every rat in the glucose group increased intake whensaccharin was added to 6?Z0glucose, whereas the mean fructoseintake showed a nonsignificant decrease when saccharin wasadded to 6Y0fructose (compared to the previous day, differencebetween groups in change scores p = 0.02). However, when thesugar concentration was lowered to 3Y0,saccharin had similareffects on intake of fructose and glucose (see Fig. 5). Every ratin the fructose group drank more 3~ofructose plus saccharin thanit did of 6Y0fructose plus saccharin, whereas this did not occurfor the glucose group (analysis of variance interaction of groupsvs. sugar level on intake of saccharin plus sugar p = 0.0004).

DISCUSSION

The results of this experiment begin to establish a consistentpicture for single bottle intake tests. Rats ingest similar amountsof 6Y0fructose and 6Y0glucose as long as saccharin is not present.The addition of 0.25% saccharin to these solutions caused rats to

+- Fructose – ● - Glucose

100

80[

0 1 2 3 4

zo~6 7 8 9 10

100 -

80 -

<60 -

41.-- --

40 -

t20L ,

13 14 15 16 17

Days

FIG. 5. Mean 3 SEM 24-h intake of sugar solutions in Experiment 6.The top, middle, and bottom panels depict the first, second, and thirdweeks of the experiment, respectively. The sugar concentration was 6Y0for the top 2 panels and 3% for the bottom panel. The solutions contained0.25% saccharin on days 9 and 16.

FRUCTOSE AND GLUCOSE 1303

increase glucose intake but not fructose intake, thereby creatinga difference between rats offered glucose and fructose. The dif-ference in outcome between Experiment 1 compared to Experi-ments 3 and 4 are apparently due to the use of saccharin in Ex-periment 1.

EXPERIMENT 7

One possible explanation for the observed interaction be-tween type of sugar and saccharin is that fructose might tastequalitatively different from glucose. Experiment 5 providedsome support for this idea, but the results for the rats havingprevious experience with fructose were ambiguous. It was un-clear whether or not rats having prior experience with fructoseactually preferred glucose over fructose. This experiment alsoattempted to determine if rats prefer fructose with saccharinover plain fructose.

METHOD

The rats were the same ones used in Experiment 6. As inExperiment 5, the rats were given 6% fructose or glucose forbrief periods. However, a few of the rats failed to drink detectableamounts of fluid during initial tests. The rats were therefore givena total of 9 brief single-bottle tests. During the first 6 tests, thesewere as long as 2 h; solutions were offered to the rats until theydrank detectable amounts of fluid. In subsequent sessions, theywere gradually reduced to 10 min. During these tests, the ratswere offered the same sugar that they drank in Experiment 6.After the 9th brief single-bottle test, the rats were rested for 2days and then offered a choice of 6% fructose vs. 6% glucosefor 10 min on 3 consecutive days. The left-right positions offructose and glucose were reversed on consecutive days. Finally,the rats were offered a choice of 6% fmctose with 0.25~0 sac-charin vs. 6% fructose without saccharin.

RESULTS

During the last 2 single-bottle tests, the fructose group drank4.2 ~ 0,7 and 5.1 ~ 0.9 g, which is not significantly differentfrom the amount drunk by the glucose group, 4.6 * 1.2 and 5.9f 1.3 g (see Fig. 6).

The first time the rats were offered a choice of glucose vs.fructose, the rats in the fructose group drank slightly more fruc-tose than gIucose, whereas the rats in the glucose group drankless fructose than glucose (interaction of sugar by groups p =0.01, see Fig. 6). This interaction was not significant on subse-quent trials (p > 0.2). Rats in the glucose group drank signifi-cantly more glucose than fructose on the first and third trials (p= 0.03), whereas rats in the fructose group never drank signifi-cantly more of one sugar than of the other (all p > 0.2).

When all rats were offered a choice of fructose with or withoutsaccharin, both groups drank more of the saccharin-containingfluid (overallp < 0.001); this difference tended to be greater forthe fructose than for the glucose group (p = 0.054, data notshown).

Total fluid intake was similar in the 2 groups on aIl preferencetests (p > 0.2).

DISCUSSION

The main findings are consistent with those from Experiments2 and 5, especially the observation that rats having previous ex-perience with glucose preferred glucose over fructose. Rats hav-ing previous experience with fmctose, but not with glucose,showed no reliable preference. Unlike Experiment 5, rats havingprevious experience with fructose did not drink more fluid overall

_ Fructose _ Glucose

‘rI

#a #9 FG FG FG

Single–Bottle #1 #2 #3Preference Tests

FIG. 6. Mean < SEM 10-min intake of sugar solutions in Experiment 6.The leftmost bars depict the 8th and 9th single-bottle tests. The otherbars depict 3 successive preference tests.

than did rats having previous experience with glucose; the reasonfor this discrepancy is not known.

Although saccharin did not seem to elevate intake of 6Y0fruc-tose in a single-bottle test, rats strongly prefer fructose with sac-charin to fructose without saccharin, although the magnitude ofthis preference appears to be influenced by prior experience.

EXPERIMENT 8

The taste of saccharin is thought to have an aversive com-ponent, analogous to the bitter off-taste often reported by humans(9). Perhaps fructose does not mask this off-taste as well as doesglucose. The aversive effects of this off-taste are most readilyapparent during the first day of exposure and decline thereafter( 10). Thus, the observation in Experiment 6, that saccharin didnot increase intake of 6Y0fructose but did increase intake of 3~ofructose, might be an order effect rather than a concentrationeffect because the 6Y0solution was offered before the 3?10solu-tion was offered. It was therefore hypothesized that lengtheningthe duration of the test to 3 days would give rats time to overcometheir avoidance of saccharin’s off-taste.

METHOD

A new set of 16 rats was divided into fmctose and glucosegroups. The rats were offered either 6% fructose plus 0.25% sac-charin or 6~0 glucose plus 0.25~0 saccharin for 3 consecutivedays. No other fluids were available over this period. Fluids wereoffered just before the lights went out in the animal room. Intakewas recorded at the end of the first 2 h, and then daily.

RESULTS

During the first 2 h, rats given fructose plus saccharin drank3.7 ~ 0,9 g whereas the rats given glucose PIUS sacchtin ~~

8.1 ~ 1.3 g (p = 0.02). Over the entire 3-day period, the ratsgiven fructose + saccharin drank substantially less than did therats given glucose + saccharin (overall p < 0.001, for each in-dividual day p s 0.02, see Fig. 7). There was a nonsignificanttrend for the difference between the fructose and glucose groups

1304 RAMIREZ

100

80

60

40

I20

~ Fructose+ Sacc.

‘-o-- Glucose+ Sacc.

T

,

0 i 2 3

DaysFIG. 7. Mean * SEM 24-h intake of sugar solutions in Experiment 8.

to increase over days (daysby groups interaction p > 0.2, seeFig.7).

DISCUSSION

Intake wasrecorded after the first 2hbecause this was felttobe the minimum time needed to observe reliable effects withnaive rats; individual rats occasionally wait more than 1 h beforesampling novel solutions. Although 2 h provides ample time forpostingestive feedback to produce satiety, it seems unlikely thatthe differential intake could be attributed to postingestive effects.Satiety should not be a confounding factor because the satiatingeffects of fructose and glucose are probably similar (23). In the-ory, conditioning should increase intake over time, but there areno reports that conditioning can occur within 2 h. If conditioningwere an important factor, it would be expected to magnify thegroup difference over time; there was no significant trend in thisregard.

GENERAL DISCUSSION

Several features of the results are either surprising or incon-sistent with generally accepted ideas about the factors governingintake of glucose and fructose.

Given the apparent concensus that these sugars differ mainlyin intensity (fructose greater than glucose), the observation thatrats preferred glucose over fructose would seem surprising. How-ever, the consensus about fructose in rodents is based on a com-bination of extrapolation from humans and very limited data on

rodents. There are very few published studies of rat preferencefor fructose vs. glucose, and these do not show any consistentpattern (5,19,41 ). Studies of the response to these sugars in briefsingle-bottle tests generally indicate that rats drink more of dilute(0.1 Ml 1.8%) fructose than of glucose (5,8,46), but only one ofthese studies reported any statistical tests (5). Electrophysiolog-ical studies in rodents have been interpreted as indicating that thetaste of fructose is a more potent chemosensory stimulus thanglucose because fructose produces a stronger response than doesglucose in the chorda tympani ( 17,29,47) and greater superficialpetrosal nerve (25). However, because this is not the case forthe glossopharyngeal nerve (27), and because individual neuronsdiffer greatly in their relative sensitivity to these sugars (3), it isnot possible to assume a one-to-one relationship between themagnitude of the neural response and the degree of perceivedintensity.

There appear to be qualitative differences between the tastesof these sugars because the relative preference for the 2 sugarswas influenced by previous experience. Rats having extensiveprior experience with glucose preferred glucose over fructose,whereas rats having extensive prior experience with fructoseshowed no consistent preference. These experiential effectsmight be attributed to a form of flavor-postingestive condition-ing, especially because the effects of previous experience withglucose seem to depend on whether the glucose was sham-fed orreal-fed (Experiment 2). However, it would be difficult to imag-ine how such conditioning could occur if fructose and glucosehave qualitatively similar tastes to rats.

If these sugars differ solely with regard to taste intensity, onemight expect that adding saccharin to them would either maskany gustatory differences, or greatly enhance the intensity of bothsugars ( 18). Contrary to this expectation, 0.2570 saccharin failedto stimulate intake of 6% fructose, but greatly stimulated intakeof 6% glucose in 24-h single-bottle tests. It could be hypothe-sized that saccharin does not stimulate intake of 6Y0fructosebecauseratscannot tolerate large amounts of fructose. This hy-pothesis is consistent with the literature on fructose intolerance(7,15,37,49), but it is not consistent with the present series ofexperiments, especially the initial study demonstrating a differ-ence between fructose and glucose for sham-fed rats. Althoughsome fraction of the sugar ingested by sham-fed rats was prob-ably absorbed (42), the total amount of fructose ingested in Ex-periment 1 was so small that the fraction retained probably hadinconsequential physiological effects. Furthermore, increasingthe sugar concentration from 6% to 127. did not differentiallyinfluence intake in Experiment 4. Indeed, rats in both the fructoseand glucose groups ingested substantially more sugar when of-fered 12% than when offered 6% sugar (Experiment 4). In Ex-periment 8, naive rats given fructose plus saccharin drank half asmuch as rats given glucose plus saccharin in a 2-h test period.

It is difficult to account for the results by assuming that, forthe type of rats used in the present studies, the two sugars havequalitatively similar tastes, but that fructose is merely a less po-tent gustatory stimulus than glucose. Because the rats exposed tothese sugars should have been undergoing flavor-postingestive-consequence conditioning ( 11,32,34,35,44), and if the post-ingestive effects of fructose are less potent than those of glucose( 1,2,39,40), one would expect the rats to drink more glucosethan fructose in 24-h single-bottle tests. This did not occur.

The results reported here would seem reasonable, instead ofperplexing, if one assumes that, for rats, the taste of glucose is amixture of sucrose-like (i.e., sweet) and maltodextrin taste,whereas fructose has only a sucrose-like taste (28). If these twosubstances taste qualitatively different, there is no reason to ex-pect that rats would consistently prefer one over the other, or

FRUCTOSE AND GLUCOSE 1305

consistently ingest more of one than the other. The principal ob-stacle to accepting this conclusion is that most of the studiesexamining the similarity of these stimuli to rodents suggest thatthey have similar tastes in rats ( 12,30), mice (26), gerbils (20),and hamsters (30). However, evidence that sugars share somecommon attribute does not rule out the possibility that they alsopossess differences. A recent report argued that the methodsused in such studies were better suited for demonstrating sim-ilarities than for detecting differences in the presence of somesimilarity (33 ).

Recent studies indicate that humans do not discriminate be-tween these sugars (4). It is possible that this apparent speciesdifference in gustation is not real. Conceivably, rats could smell,

rather than taste, impurities present in purified sugars (36).However, because electrophysioIogical studies suggest that ratspossess different receptors for glucose and fructose (3), itseems parsimonious to assume that the present findings are dueto gustatory effects. Perhaps a better understanding of speciesdifferences in gustation could provide clues about the biologicalfunctions of sugar preference, something that is poorly under-stood (31 ) .

ACKNOWLEDGEMENTS

This work was supported by an NIH grant ( 1 RO1 DCO11O7), andthe Monell Chemical Senses Center. I thank M. Tordoff for lending mehis sham-feeding apparatus.

REFERENCES

1. Ackroff, K.; Sclafani, A. Flavor preferences conditioned by sug-ars:Rats learn to prefer glucose over fructose. Physiol. Behav.50:815-824; 1991.

2. Ackroff, K.; Schfani, A. Sucrose to Polycose preference shifts inrats: The role of taste, osmolality and the fructose moiety. Physiol.Behav. 49:1047– 1060; 1991.

3. Beidler, L. M.; Tonosaki, K. Multiple sweet receptor sites and tastetheory. In: Pfaff, D. W., ed. Taste, olfaction, and the central nervoussystem. New York: The Rockefeller University Press; 1984:47–64.

4. BresIin, P. A. S.; Kemp, S.; Beauchamp, G. K. SingIe sweetnesssignal. Nature 369:447–448; 1994.

5. Cagan, R. H.; Mailer, O. Taste of sugars: Bnefexposure single-stimulusbehavioral method. J. Comp. Physiol. Psychol. 87:47-55; 1974.

6. Campbell, C. S.; Davis, J. D. Chronic intrajugukw, intraportal, gas-tric, and duodenal cannulae for the rat. In: Singh, D., ed. Physiolog-ical techniques in behavioral research. Monterey, CA: Brooks/Cole;1975:163-177.

7, Cori, C. F. The fate of sugar in the animal body. I. The rate ofabsorption of hexoses and pentoses from the intestinal tract. J. Biol.Chem. 66:691-715; 1925.

8. Davis, J. D. The effectiveness of some sugars in stimulating lickingbehavior in the rat. Physiol. Behav. 11:39-45; 1973.

9. Dess, N. K. Saccharin’s aversive taste in rats: evidence and impli-cations. Neurosci. Biobehav. Rev. 17:359–372; 1993.

10. Domjan, M.; Gillan, D. Role of novelty in the aversion for increas-ingly concentrated saccharin solutions. Physiol. Behav. 16:537–542; 1976.

11. Drucker, D. B.; Ackroff, K.; Sclafani, A. Nutrient-conditioned flavorpreference and acceptance in rats: Effects of deprivation state andnonreinforcement. Physiol. Behav. 56:701 –707; 1994.

12. Dugas du Villard, X.; Her, C. Qualitative discrimination of sweetstimuli: behavioral study on rats. Chem. Senses 6: 143– 148; 1981.

13. Edgington, E. S. Randomization tests, 2nd ed. New York: Dekker;1987.

14. Emits, T.; Corbit, J. D. Taste as a diposegenic stimulus. J. Comp.Physiol. Psycho]. 83:27–31; 1973.

15. Fujisawa, T.; Riby, J.; Kretchmer, N. Intestinal absorption of fruc-tose in the rat. Gastroenterology 101:360–367; 1991.

16. Gilbert, R. M.; Sherman, I. P. Palatability-induced polydipsia: sac-charin, sucrose, and water intake in rats, with and without food dep-rivation. Psychol. Rep. 27:3 19–325; 1970.

17. Hardiman, C. W. Rat and hamster chemoreceptor responses to alarge number of compounds and the formulation of a generalizedchemosensory equation. Ph.D. Dissertation, Florida State Univer-sity. Ann Arbor, MI: University Microfilms, Inc. #64-t0,585; 1964.

18. Hyvonen, L.; Kurkela, R.; Koivistoinen, P.; Ratilainen, A. Fructose-saccharin and xylitol-saccharin synergism. J. Food Sci. 43:251 –254;1978.

19. Jacobs, H. L. Taste and the role of experience in the regulation offood intake. In: Kare, M. R.; Mailer, O., eds. The chemical sensesand nutrition. Baltimore: Johns Hopkins Press; 1967:187–200.

20. Jakinovich, W. J. Taste aversion to sugars by the gerbil. Physiol.Behav. 1982:1065– 1071; 1982.

21. Jakinovich, W. Jr.; Sugarrnan, D. Sugar taste reception in mammals.Chem. Senses 13:13-31; 1988.

22. Kissileff, H. R. Manipulation of the oral and gastric environments.In: Myers, R. D., ed. Methods in psychobiology, Volume 2. NewYork: Academic Press; 1972:125– 153.

23. Maggie, C. A.; Decarr, L. B.; Vasselli, J. R. Differential effects ofsugars and the alpha-glucosidase inhibitor acarbose (Bay G 5421 )on satiety in the Zucker obese rat. Int. J. Obesity 11 (Suppl. 3):53–56; 1987.

24. Manly, B. F. J. Randomization and monte carlo methods in biology.London: Chapman & Hall; 1991.

25. Nejad, M. S, The neural activities of the greater superficial petrosalnerve of the rat in response to chemical stimulation of the palate.Chem. Senses 11:283–293; 1986.

26. Ninomiya, Y.; Higashi, T.; Katsukawa, H.; Mizukoshi, T.; Funa-koshi, M. Qualitative discrimination of gustatory stimuli in threedifferent strains of mice. Brain Res. 322:83–92; 1984.

27. Ninomiya, Y.; Kajiura, H.; Mochizuki, K. Differential taste re-sponses of mouse chorda tympani and glossopharyngeal nerves tosugars and amino acids. Neurosci. Lett. 163:197–200; 1993.

28. Nissenbaum, J. W.; Sclafani, A. Qualitative differences in polysac-charide and sugar tastes in the rat: a two-carbohydrate taste model.Neut-osci. Biobehav. Rev. 11:187–196; 1987.

29. Noma, A.; Goto, J.; Sate, M. The relative taste effectiveness of var-ious sugars and sugar alcohols for the rat. Kumamoto Med. J. 24: 1–9; 1971.

30. Nowlis, G. H.; Frank, M. E.; Pfaffmann, C. Specificity of acquiredaversions to taste qualities in hamsters and rats. J. Comp. Physiol.Psychol. 94:932–942; 1980.

31. Ramirez, I. Why do sugars taste good? Neurosci. Biobehav. Rev.14:125–134; 1990.

32. Ramirez, I. Stimulation of fluid intake by carbohydrates: Interactionof taste and calories. Am. J. Physiol. 266:R682–R687; 1994.

33. Ramirez, 1. Sucrose and fructose have qualitatively different flavorsin rats. Physiol. Behav. 56:549–554; 1994.

34. Ramirez, 1. Stimulation of fluid intake by maltodextnns and starch.Physiol. Behav. 57:687–692; 1995.

35. Ramirez, L Stimulus specificity in flavor acceptance learning. Phys-iol. Behav. 60:595–610; 1996.

36. Rhinehart-Doty, J. A.; Schumm, J.; Smith, J. C.; Smith, G. P. Anontaste cue of sucrose in short-term taste tests in rats. Chem. Senses19:425–431; 1994.

37. Rumessen, J. J.; Gumand-Hoyer, E. Absorption capacity of fructosein healthy adults. Comparison with sucrose and its constituent mono-saccharides. Gut 27: 1161– 1168; 1986.

38. Sclafani, A. Carbohydrate taste, appetite, and obesity: An overview.Neurosci. Biobehav. Rev. 11:131– 153; 1987.

39. Sclafani, A.; Ackroff, K. Glucose- and fructose-conditioned flavorpreferences in rats: Taste vs. postingestive conditioning. Physiol.Behav. 56:399–405; 1994.

40. Sclafani, A.; Cardieri, C.; Tucker, K.; Blusk, D.; Ackroff, K. Intra-gastric glucose but not fructose conditions robust flavor preferencesin rats. Am. J. Physiol. 265:R320–R325; 1993.

1306 RAMIREZ

41.

42.

43.

44.

45.

Sclafani, A.; Mann, S. Carbohydrate taste preferences inrats:glucose, sucrose, maltose, fructose and Polycose compared.Physiol. Behav. 40:563-568; 1987.Sclafani, A.; Nissenbaum, J. W. Is gastric sham-feeding really sham-feeding? Am. J. Physiol. 248:R387-R390; 1985.Sclafani, A.; Nissenbaum, J. W. On the role of the mouth and gutin the control of saccharin and sugar intake:a reexamination of thesham-feeding preparation. Brain Res. Bull. 14:569–576; 1985.Sclafani, A.; Nissenbaum, J. W. Robust conditioned flavor prefer-ence produced by intragastric starch infusions in rats. Am. J. Physiol.255:R672-R675; 1988.Sclafani, A.; Nissenbaum, J. W.; Ackroff, K. Learned preferencesfor real-fed and sham-fed Polycose in rats: Interaction of taste, post-ingestive reinforcement, and satiety. Physiol. Behav. 56:331–337;1994.

46.

47.

48.

49.

50.

Smith, J. C.; Davis, J. D.; O’Keefe, G. B. Lack of an order effect inbrief contact taste tests with closely spaced test trials. Physiol. Be-hav. 52:1107–1111; 1992.Tonosaki, K.; Beidler, L. Sugar best single corda tympani nerve fiberresponses to various sugar stimuli in rat and hamster. Comp.Biochem. Physiol. 94A:603-605; 1989.Tordoff, M. G.; Ulrich, P. M.; Srmdler, F. Flavor preferences andfructose: evidence that the liver detects the unconditioned stimulusfor calorie-based learning. Appetite 14:29-44; 1990.Tmswell, A. S.; Seach, J. M.; Thorbum, A. W. Incomplete absorp-tion of pure fructose in healthy subjects and the facilitating effect ofglucose. Am. J. Clin. Nutr. 48:1424–1430; 1988.Van Vort, W.; Smith, G. P. The relationship between the positivereinforcing and satiating effects of a meal in the rat. Physiol. Behav.30:279-284; 1983.