Sucrose Intolerance in Birds: Simple Nonlethal Diagnostic Methods and Consequences for Assimilation of Complex Carbohydrates

The Auk 111(1):170-177, 1994

SUCROSE INTOLERANCE IN BIRDS: SIMPLE NONLETHAL DIAGNOSTIC METHODS AND CONSEQUENCES FOR ASSIMILATION OF COMPLEX CARBOHYDRATES

HYLARY L. MALCARNEY, CARLOS MARTINEZ DEL RIO, AND VICTOR APANIUS Departmentof Ecologyand EvolutionaryBiology, Princeton University, Princeton,New Jersey08544, USA

ABSTR^CT.--GrayCatbirds (Dumetella carolinensis, Mimidae) and Purple-headedGlossy-Star- lings (Larnprotornispurpureiceps, Sturnidae) showed depressed ingestion and increasedfecal sugar contentswhen shifted from glucoseand fructoseto sucrosesolutions. These species also exhibited no increasesin plasmaglucose after ingestionof sucrose,but an increasein plasmaglucose after ingestionof equicaloricdoses of a mixtureof glucoseand fructose.In vitromeasurements of intestinal disaccharidaseactivities in D. carolinensisrevealed insignif- icantsucrase activity, and low levelsof maltaseactivities. These results support the hypothesis that sucroseintolerance is a shared-derivedcharacter of the monophyleticlineage that includes starlings, mimids, and thrushes, and indicate that sucroseintolerance in birds can be easilydiagnosed with a combinationof behavioraland nonlethalphysiological measurements. We suggestthat, in birds, low intestinal maltaseactivity is correlatedwith the lack of sucraseactivity. We further hypothesizethat sucrose-intolerantbirds are poor at assimi- lating complexcarbohydrates. Received I March 1993, accepted 17 November1993.

PROXIMATENUTRIENT ANALYSIS distinguishes carbohydrate compositions.Nectar and fruit between two broad classesof carbohydrates: pulp contain sucrose,glucose, and fructose in structuralcarbohydrates, largely composedof varying proportions(Baker and Baker 1983,Ba- fi-l,4 polysaccharideslike cellulose,and soluble ker et al. 1993). The chemical differences be- carbohydratesincluding toorio- and disaccha- tween thesesugars are relatively small. Sucrose rides and a-l,4 and a-l,6 polysaccharideslike is a disaccharideof glucoseand fructose,and starch,amylopectin, and glycogen(Whistler and the monosaccharidesglucose and fructosediffer Daniel 1985). Most vertebrateshave endoge- only in the position of the carbonyl group. To nousenzymes that canhydrolyze the a linkages be assimilated, sucrosehas first to be hydro- in starchand glycogen,but do not possessen- lyzed into its components--glucoseand fruc- zymes capable of breaking the fi linkages of rose--by the intestinal enzyme sucrase.Glucose cellulose and hemicellulose (Vonk and Western and fructoseare absorbeddirectly by the intes- 1984). Vertebrates that assimilate the structural tine (Alpers 1987). Some frugivorous birds lack carbohydratecomponent in food have to rely intestinal sucraseactivity and cannot split the on time consuming microbial fermentation sucrosebond between glucose and fructose(e.g. (Stevens 1988). Structural carbohydratesare Martinez del Rio and Stevens 1989). For these generally considered "hard" to assimilate species,sucrose is a uselessenergy source that whereassoluble carbohydratesare considered can cause osmotic diarrhea and a consequent "easy" to assimilate (Prop and Vulink 1992). feeding aversion (Brugger and Nelms 1991). Here we presentdata showing that this pattern Thesesame sucrose-intolerant birds avidly in- has exceptions and that some soluble carbo- gest and profit from glucoseand fructose. hydratesmay be extremelydifficult or impos- Martinez del Rio (1990)reported that sucrase sible to assimilateby a broad group of birds. is lacking in severalspecies of birds in the fam- Martinez del Rio et al. (1992) demonstrated ilies Sturnidae (starlings) and Muscicapidae that even subtle differences in the chemical (thrushes).DNA-DNA hybridization data sug- structureof solublecarbohydrates can result in gestthat thesefamilies are part of a monophy- differences in bird digestive-utilization effi- letic lineage (Sibley and Ahlquist 1990).These ciency,in preferencesamong these substances, dataalso suggest that membersof the Mimidae and presumablyin differential utilization by (catbirds,mockingbirds and thrashers)are the birdsof foodplants containing different soluble closestliving relatives of starlings(Sibley and

170 January1994] SucroseIntolerance in Birds 171

Ahlquist 1984, 1990). Martinez del Rio (1990) cose. This latter method is a modification of a test hypothesized that lack of sucraseis a shared- commonly used to diagnosedisaccharide intolerance derived trait of the monophyletic lineage that in humans (Isokosi et al. 1972, Krasilnikoff et al. 1975). includes starlings, thrushes, and catbirds (the We validated thesemethods in D. carolinensisby measuring the activity of intestinal disaccharides(includ- sturnid-muscicapidlineage sensu Sibley and ing sucrase)in vitro. Ahlquist 1990). Here we explore behavioral and Dumetella carolinensis(mean mass 36.3 + SD of physiological correlatesof sucroseingestion in 4.0 g, n - 9) are among the most frugivorous birds in two speciesof fruit-eating birds in this lineage: temperate North America (Martin et al. 1951, Bent Gray Catbirds (Dumetellacarolinensis, Mimidae) 1948).The fractionof fruit in Gray Catbird diets can and Purple-headed Glossy-Starlings(Lampro- be as high as 95% during the fall and winter (Beal tornispurpureiceps, Sturnidae). Based on the phy- 1897,Blake and Loiselle 1992),and is reducedonly logeneticaffinities of Gray Catbirdsand Purple- during the breeding seasonwhen they becomemain- headedGlossy-Starlings, we predictedthat they ly insectivorous(White and Stiles 1990, Helmy and would: (1) lack intestinalsucrase activity; (2) be Martinez del Rio unpubl. data). Lamprotornispurpurei- ceps(mean mass= 79.4 + 10.6g, n = 4) are gregarious unable to assimilatesucrose; and (3) reject su- forest birds widely distributed in West and Central crosein feeding trials. Africa (Nigeria, Gabon,Central African Republic,and Maltose and isomaltose are the most abun- Uganda;Mackworth-Praed and Grant 1955,Sibley and dant products of the hydrolysis of starch by Monroe 1990). Their diet apparently also consists salivaryand pancreaticamylases. Maltose is hy- largely of fruit (Beecher 1978, Serle et al. 1977). drolyzedby two independentenzyme systems: We captured eight immature D. carolinensiswith sucrase-isomaltaseand maltase-glucoamylase mist nets in early October 1991 at the Hutchinson (Semenza and Auricchio 1989). Becausesucrase- Memorial Forest,New Jersey,and at a secondaryfield isomaltaseis a powerful maltase,its deficiency adjacent to the Sourland Mountain State Park, New in humans is often associated with reduced abil- Jersey(birds aged following Suthersand Suthers1990). Four L. purpureiceps(two malesand two females)were ity to utilize dietary starch(Auricchio et al. 1963, obtained on loan from the Bronx Zoo. Birds were 1972). Isomaltaseis hydrolyzed by sucrase-iso- individually housedin 48 x 48 x 48 cm cagesat 21øC maltase,and to a small degree by maltase-glu- and on a 12L/12D daily cycle. Food and water were coamylase.Martinez del Rio (1990) suggested supplied ad libitumexcept during experiments. Birds that birds lacking intestinal sucrasealso would were fed a mixed diet of Ziegler soft-billed bird diet be poor hydrolyzers of maltose.We predicted (Ziegler Bros. Inc, Gardners,Pennsylvania) and ba- that, if sucraseactivity were low or missingin nana mash (a mixture of mashed ripe bananas, veg- D. carolinensis,both maltase and isomaltase ac- etableoil, soyprotein isolatecomplemented with me- tivity would be low. thionine, and a vitamin supplement in an agar-based gel; Denslow et al. 1987). All experiments were con- MATERIALS AND METHODS ducted from October 1991 to February 1992. At the end of the experiments,L. purpureicepsindividuals Sucrose intolerance in vertebrates is most com- were returned to the Bronx Zoo. The surviving D. monly causedby the absenceof intestinal sucrase carolinensisindividuals were releasedin August 1992. activity (Gudman-Hoyer et al. 1984). Measuring su- We conducted all behavioral tests at the onset of crase activity requires biochemical analysis of intes- the light period (0800 EST). In each trial we removed tinal tissue (Dahlqvist 1984). Obtaining samples of food and water, and lined the bottom of the cages intestinaltissues from live smallbirds is not yet prac- with teflon-coatedplastic to facilitate excreta collec- tical and, consequently, measuring sucraseactivity tion. Each test consisted of four 2-h trials conducted requires sacrificingbirds and extracting the intestine in four successivedays. We offered birds sugar so- (Martinez del Rio 1990). To diagnose sucroseintol- lutions in glass tubes consisting of an upper 42-cm erance in D. carolinensisand L. purpureiceps,we used section and a lower 7-cm section bent upward at a 45ø two nonlethaland minimally invasivemethods. The angle. Birds drank from an elliptical hole (2 x 1.5 first method relies on sequentially feeding birds a 1:1 cm) at the distal end of the lower portion of the tube. mixture of glucose and fructose, and then offering In the first two trials we presented birds with a tube them sucrose. Sucrose-intolerant birds should show containing a 1:1 mixture of glucoseand fructose(15% depressedingestion and increasedfecal sugar con- mass/totalvolume). On days3 and 4, birds were pre- tents when shifted from glucoseand fructoseto su- sented with an equicaloric solution of sucrose.At the crose. The second method involves measuring the end of each trial we measured the amount of test increasein plasmaglucose after birds have been chal- solution consumed and collected 5 to 10 samples of lenged with an oral dose of sucrose.Sucrose-intol- excreta from the bottom of each cage. We measured erant birds should show nil increasesin plasmaglu- fecal sugar from these samples with a temperature 172 MALCARNEY,MARTINEZ DEL RIO, AND APANIUS [Auk,Vol. 111 compensatedrefractometer (Reichter-Jung10431). 250 ml 1.0 M Tris/HC1, pH 7, plus 250 ml 0.5 Na- Although uratesand fecal materials can make refrac- H2PO;/Na2HPO4,pH 7). Glucosestandards (0-120 •g tometer readings irmacurate, the refractive index of in 200 •1 0.1 M sodium maleate buffer, pH 7) were bird fecal samplesprovides a consistentrelative in- reacted with the stop develop reagent to obtain a dicatorof sugarexcretion (Martinez del Rio et al. 1989, standard curve. Protein concentration in tissue ho- Bruggerand Nelms 1991).We report fecal sugarcon- mogenateswas measuredusing the Bio-Radkit (Bio- centration in percent of sucrose(Brix = massof sugar Rad, Richmond, California) with bovine serum al- per volume of solution; Bolten et al. 1979). bumin standards.To allow comparisonwith other We measuredthe responsein plasmaglucose levels intestinal disaccharidase studies we calculated disac- (PGL) in birdssubject to two treatments:a 1:1glucose: charidaseactivities using three different standardiza- fructose intubation, and a sucrose intubation. Birds tion procedures:activity per unit intestinal area as were fasted overnight and intubated with 3 g/kg of •mol.(min) '.(cm 2nominal area) •; activity per gram a 15%(mass/volume) sugar solution. After 30 min we of protein as •mol.(min) '.(cm2.gram of protein) '; obtained approximately 200 •1 of blood by jugular and total hydrolyric capacityas •mol.(min) • venipuncture using a 0.5-ml syringe with a 28-gauge Statisticalanalysis.--Although we report descriptive needle (Hoysak and Weatherhead 1991).We obtained statisticsfor both speciesin all experiments,we only a blood sampleafter 30 min becausepreliminary data report significance values from inferential statistics indicated that plasmaglucose peaks between 25 and testsfor data on D. carolinensisfor which samplesize 35 min after a 1:1 glucose:fructose challenge (see was adequate (eight individuals). The number of in- Martinez del Rio et al. 1988:fig. 3). Bloodwas trans- dividuals of L. purpureicepsstudied was too small (four ferred to chilled heparinizedmicrocapillary tubes and individuals) to allow use of inferential statistics. For centrifuged for 5 min (IEC microhematocrit centri- paired data we usedsign tests,which are robust albeit fuge; see Cohen 1966).We measuredplasma glucose conservative (Mosteller and Tukey 1977). To estimate after color development (Glucose-Trinder 500 re- parametersof linear regressionswe used standard agent, Sigma Chemical Co.) on a spectrophotometer least-squaresmethods. Results are given as œ -+ SD. (Beckman DU-64) set at 505 nm. As an estimate of the responseto a glucose:fructose or sucrosechallenge RESULTS we used the difference in PGL 30 rain after treatment and PGL in untreated birds fastedovernight. Because Neither D. carolinensisnor L. purpureiceps we were concernedabout stressingbirds excessively changed consumption of glucose and fructose by repeatedly drawing blood in a short time, we ob- solution from day 1 to day 2 (P > 0.I; Fig. I). tained a single fasting plasmameasurement per individual. This procedureis justified becausein birds Consumption decreased in all individuals of fasting PGL remains relatively constant at the tem- both speciesfrom day 2 to day 3, when birds poral scaleof our experiments(three weeks;Martinez were shifted from glucoseand fructoseto su- del Rio and Phillips unpubl. data).Plasma osmolarity crose (P < 0.05). This decline in consumption was measured on a Wescor 5500 vapor pressureos- continued from day 3 to day 4 in all birds but mometer. one glossy-starling,which drank very little of Three D. carolinensis individuals were euthanized the sucrose solution during either days (P < with a halothane overdose. The small intestine was 0.05, Fig. 1). Fecal sugar levels over the four- immediately excised,divided into three sectionsof day trials followed a similar pattern in both equal length, and placed in ice-cold saline (1.02%). species(Fig. I). Fecalsugar remained low (<2% We slit eachsection longitudinally and measuredits length and width to obtain an estimate of intestinal BRIX) the first two days of the trials (Fig. I), nominal area.After weighing, eachsection was stored increasedon day 3 when birds were shifted to in liquid nitrogen. We measuredsucrase, maltase, and sucrosesolutions (P < 0.05),and remained high isomaltaseactivities in theseintestinal samplesusing through day 4 (P > 0.I, Fig. 1). These results a previouslydescribed method (Martinez del Rio et suggesthigh assimilation of glucose and fruc- al. 1988)from Dahlqvist(1984). Martinez del Rio (1990) tose, but low assimilation of sucrose. provided details of the disaccharideassay. Briefly, We found the extremely broad range of con- tissueswere homogenized in 350 mM mannitol in 1 centrationsof fastingplasma glucose that seems mM Hepes/KOH, pH 7.5 (30 s at OMNI 5000 homoge- to be typical of bird species(Hazelwood 1984, nizer setting6), and tissuehomogenates (100 •1) were Groscolasand Rodriguez 1981,Marsh et al. 1984). incubatedat 40øC(Prinzinger et al. 1991) with 100 •1 of 56 mM sugar (sucrose, maltose, and isomaltase) Fastingplasma glucose ranged from 260 to 335 solutionsin 0.1 M maleate/NaOH buffer, pH 6.5 for rag/100 ml in L. purpureiceps,and from 287 to 10 min. After incubation, reactionswere arrestedby 458 rag/100 ml in D. carolinensis(Fig. 2). All adding 3 ml of a stop/develop reagent (one bottle of individuals showed increased PGL relative to Glucose-Trinder500 reagent[Sigma Chemical Co.] in fasting levels 30 min after administration of January 1994] SucroseIntolerance in Birds 173

20 20- 15./• Days1and 2:glucose + fructose 15• s3 and4 : s •

10. 10 •

5. 5

O- 0 1 2 3 4 1 2 3 4

Days Days Fig. 1. Ingestionof glucose:fructose (days 1 and 2) and sucrose(days 3 and 4) solutions(open symbols) and feca!sugar concentration after ingestion(closed symbols) in D. carolinensis(circles) and L. purpureiceps individuals(triangles). Solutions of glucose:fructose and sucrosewere equicaioric(15% mass/totalvolume). glucoseand fructose(Fig. 2; P < 0.05). The mean increasesabove fasting levels in L. purpureiceps and D. carolinensis were 104.9 + 19.2 and 191 +

18.2 rag/100 ml, respectively.In D. carolinensis 542 the increase in PGL over fasting levels after ß Y=X administration with glucoseand fructose was 492' linearly and negatively correlatedwith fasting PGL (Y = 350.9 - 0.7X; r = 0.77, P < 0.005), ß ß suggestingregulation of maximal plasma glu- coseconcentration. As expected for sucrose-intolerant animals, none of the birds intubated 392' with sucrose showed increased PGL (Fig. 2). Surprisingly, all D. carolinensisindividuals 342 ' showed a small (-18.4 ñ 3.2 mg/100 ml) but significant (sign test, P < 0.05) decreasein PGL 30 min after intubation with sucrose(Fig. 2). 292 ' OA•00 Plasma osmolarity did not vary significantly øø among treatments (P > 0.05). Mean plasma os- 242' A molarities for L. purpureicepsand D. carolinensis were 385.2 + 28.3 and 350.8 + 8.4 mg/100 ml, 2;8 3•8 3;8 4•8 45'8 respectively. FastingPlasma Glucose (mg/100 ml) We found maltase and isomaltase activities in three D. carolinensisindividuals, but only traces Fig. 2. Plasmaglucose concentration after 30 min of sucraseactivity in two individuals and no of glucose:fructose (closed symbols) and sucrose (open detectablesucrase activity in another (Table 1). symbols)oral doses(3 g/kg in a 15%solution). Circles Sucraseactivity was extremely low (less than representD. carolinensisand trianglesL. purpureiceps individuals.For D. carolinensis,mean plasma glucose 0.001 g of sucrose hydrolyzed-h '.individu- 30 min after a glucose:fructose dose was 481 + 14 al •). Martinez del Rio (1990)reported that mal- mg/100 ml and is representedby horizontal dotted tasewas linearly and positively correlatedwith line. Identicalline (Y = X) denotingno posttreatment sucraseactivity in a sampleof 11 speciesof pas- changein PGL relativeto fastinglevels is shownas serine birds. As predicted, mean maltase activ- reference. 174 MALCARNEY,MARTINEZ DEL RIO, AND APANIUS [Auk,Vol. 111

TABLE1. Intestinal disaccharidaseactivities in three individuals of Dumetellacarolinensis (œ ñ SD). Disac- charidasesmeasured by glucoseliberated after incubatingtissue homogenates with 0.28 M substrateat 40øC for 10 min.

Total activity Activity/area Activity/mg of protein (•mol.min •) (•mol.min •-cm 2) (•mol.min t.g of protein Maltase 28.96 ñ 14.12 2.38 + 0.88 102.03 _+ 31.03 Isomaltase 0.29 + 0.12 0.02 ñ 0.01 2.77 + 1.81 Sucrase • 0.04 ñ 0.05 0.004 + 0.001 0.50 + 0.43

Only two individuals showeddetectable sucrase activity. ity standardized per unit nominal area of in- purpureicepsand D. carolinensisare sucrose intestine ([total maltose hydrolysis]/[total small tolerant, and indicate that in D. carolinensis this intestine normal area]) in D. carolinensiswas low intoleranceis causedby the absenceof signif- and within the 95% confidence interval of the icant intestinal sucraseactivity. The behavioral, intercept of the maltase-versus-sucraseregres- fecal sugar, and blood testsappear to be easy sion line for these 11 species(Fig. 3). nonlethaltechniques to diagnosepoor sucrose digestion in birds. Care must be exercised in DISCUSSION the interpretation of behavioral testsand fecal sugar tests,however. Birds that possesssucrase In analogywith human clinical terminology, activity but that are relatively inefficientat di- we define "sucrose intolerance" in birds as a gesting sucrose,such as Cedar Waxwings (Bom- combination of symptoms including sucrose bycillacedrorum), can show sucroseaversion and realabsorptionas evidenced by high fecalsugar increasedfecal sugar concentrationwhen ex- concentration, aversion to sucrosein sequential posedto sucrose-containingfood (Martinez del feeding trials, and a "flat" responsein plasma Rio et al. 1989,K. E. Bruggerunpubl. data).Ap- glucoseafter sucroseingestion (Gudman-Hoyer parently, a reduction in consumption after ex- et al. 1984). Our results demonstrate that L. posure to sucroseaccompanied by increased fecal sugar concentration can be reliably interpreted ascaused by inefficient digestionof sucrose,but not necessarilyas evidence for lack of intestinal sucraseactivity. Table 2 summa- • 16 rizes sugar fecal output in five speciesof sucrose-fedbirds, including Cedar Waxwings.Fe- cal sugarin thesespecies varies relatively little and does not differ significantly between spe- • 10 o cies lacking sucraseand Cedar Waxwings = 1.65, P > 0.2), which digest sucroseineffi- ciently (Martinez del Rio et al. 1989). Why do thesefive speciesexhibit suchsimilar fecal concentrations when fed on sucrose? As- sumingthat all fecal solutesare undigestedsu- crose and transforming BRIX% to osmolarity yields a mean fecal osmolarity equal to 380 _+ SucraseActivity (]amol .min '•. cm '2) 15 raM. This value is slightly higher than the averagevalue for passerineplasma osmolarity Fig. 3. Linear regression (thick line) and 95% con- (342 + 18.5 raM; Skadhauge1981). Birds with fidenceintervals (thin lines) for relationshipbetween poor or no digestion of sucroseare apparently intestinal sucraseand maltaseactivities in 11 species of passerinebirds (open circles).Relationship exclud- incapable of concentratingexcreta against a ing D. carolinensisis Y = 1.4 + 7.4X; r = 0.95; Martinez concentration gradient to osmolarities much del Rio 1990).Closed symbol represents average val- higher than plasma(see Skadhauge1981:92). ues for D. carolinensis(from Table 1). Error bars are Becauseplasma osmolarity is very similaramong SE. Note that point for D. carolinensisis within the small passerinespecies, fecal sugar after sucrose 95% confidenceinterval for passefineregression. ingestionshould also be similar. A corollaryof January1994] SucroseIntolerance in Birds 175

TABLE2. Comparison of fecal sugar in sucrose-fedGray Catbirds (Dumetellacarolinensis), Purple-headed Glossy-Starlings(Larnprotornis purpureiceps), European Starlings (Sturnus vulgaris), American Robins(Turdus migratorius),and Cedar Waxwings (Bombycillacedrorum). The latter speciesexhibits sucraseactivity, but is relatively inefficientat digestingsucrose (see Martinez del Rio et al. 1989).

Species Fecal sugar• Reference Dumetella carolinensisb 14.5 ñ 3.4 (8) This report Lamprotornispurpureiceps b 11.6 _+2.9 (4) This report Sturnusvulgaris 11.5 ñ 2.1 (7) Brugger et al. (1992) Turdusmigratorius 11.6 _+1.1 (4) Brugger and Nelms (1991) Bombycillacedrorum 12.1 _+2.6 (10) Martinez del Rio et al. (1989)

% Brix +- SD (n). Measurements on day 3 (see text). this hypothesis is that sucrose-intolerantbirds imal possible contribution of maltose to the should suffer from net water losswhen feeding energy intake of D. carolinensis.In our calcula- on concentrated sucrose solutions. Indeed, tion we assumedthat luminal digestion of car- American Robins(Turdus migratorius) fed on su- bohydrates into oligosaccharidesby pancreatic crose dramatically increased water consump- amylaseswas not limiting. If maltose concen- tion (Bruggerand Nelms 1991).In addition, fe- tration is at saturating concentration in the gut cal sugarconcentration in sucrose-fedEuropean (Kinfor intestinal maltose hydrolysis in D. car- Starlings(Sturnus vulgaris) is approximatelythe olinensisis about 3 mM; Martinez del Rio un- same irrespectiveof the sucroseconcentration publ. data) then D. carolinensisindividuals can in food (ca. 11% BRIX; Brugger et al. 1993). hydrolyze 0.6 + 0.1 g maltose/h, which pro- Plasmaglucose tests also supported the hy- vides 9.6 + 1.6 kJ/h (assuming16 kJ/g of mal- pothesisthat D. carolinensisand L. purpureiceps tose;Weast and Selby 1967).The predictedfield are sucrose intolerant. Although fasting PGL metabolic rate of a D. carolinensis individual was variable, the negative correlation between weighing 36.3 g is 131 kJ/day (Nagy 1987). Dur- fasting PGL and the increase in PGL after a ing a 12-h day, catbirdscan hydrolyze a maxi- glucose:fructose challenge indicates that in D. mum of about 115 kJ maltose/day, or 88% of carolinensisa maximal PGL of about 536 mg/100 their metabolic needs. Dumetella carolinensis in- ml is defended (Fig. 2). We have no adequate dividuals apparently cannot meet their ener- explanationfor why PGL showed a significant getic demands from maltose alone. Their ca- decreaserelative to fasting levels after a sucrose pacity to hydrolyze maltoseis less than that challenge. required to fuel metabolism. Measurementson In vitro measurements indicated that sucrose other vertebrate speciesthat possessintestinal intolerance in D. carolinensis seems to be caused sucraseactivity indicatemaltase hydrolytic abil- by lack of intestinal sucraseactivity. The mag- ities that are several times higher than those nitude of intestinal sucraseactivity detected in needed to fuel metabolism on a maltose (or D. carolinensisin vitro was too low to be of phys- starch) diet (Hernfindez and Martinez del Rio iological significance and similar to that re- 1992). ported by Martinez del Rio (1990) for two spe- Our resultsalso provide support for the hy- cies of thrushes (Muscicapidae). Intestinal pothesisthat sucroseintolerance is a shared-de- maltase and isomaltasewere present, albeit at rived characterof the sturnid-muscicapidlineage low levels. Mean maltase activity in D. caroli- (Martinez del Rio 1990). Although the reasons nensiswas within the limits predicted for a bird why sucraseactivity was lost and why sucrasehas lacking sucraseactivity, supporting the hy- not been regained in the ancestorof starlings pothesisthat birds unable to digestsucrose also and thrushes are unclear, the consequencesof have reduced maltosehydrolyzing abilities.The sucraseabsence in this lineage seem to be im- available data on isomaltaseactivity in birds is portant. Lack of sucraseactivity seemsto limit still too scanty to allow comparative conclu- the dietary choicesof a large number of species sions. (Martinez del Rio and Stevens 1989) and also To evaluate the possibleconsequences of re- probably plays a significant role in the inter- duced maltose hydrolysis for birds lacking in- action of birds with the plants whose seedsthey testinal sucraseactivity, we calculatedthe max- disperse.Sucrose intolerance in starlings,mim- 176 MALCARNEY,MARTINEZ DEL RIO, AND APANIUS [Auk,Vol. 111 ids and thrushesappears to be a strongselective lution in the starlings.Bull. Chicago Acad. Sci. force that contributes to the maintenance of low 11:269-298. sucroseconcentrations in fruit pulp and to the BENT, A. C. 1948. Life histories of north American prevalenceof glucoseand fructosein present- nuthatches, wrens, thrashers, and their allies. Bull. U.S. Natl. Mus. 195. day bird-dispersedplants (Martinez del Rio et BLAKE,J. G., AND B. A. LOISELLE. 1992. Fruits in the al. 1992). Sucroseintolerance in birds may be a diets of Neotropical migrants in CostaRica. Bio- good example of the implicationsthat single tropica 24:200-210. and seeminglytrivial evolutionary events,such BOLTEN,A. B., P. FEINSINGER,H. G. BAKER,AND I. BA- asthe lossof activity in a single multifunctional KER. 1979. On the calculationof sugar concen- enzyme, can have on the interaction between tration in flower nectar. Oecologia 41:301-304. animals and plants. BRUGGER,K. E., AND C. O. NELMS. 1991. Sucrose avoidanceby American Robins (Turdusmigrato-

ACKNOWLEDGMENTS rius):Implications for control of bird damagein fruit crops.Crop Protection10:455-460. This manuscript is the result of a collective effort. BRUGGER,K. E., P. NOL, AND C. I. PHILLIPS. 1993. Re- We are grateful to those that participated in its com- pellency of sucroseto European Starlings:Will pletion. H. Suthersand E. Stileshelped in the capture high-sucrosecultivars deter bird damagein fruit of catbirdsand in conditioning them for releaseafter crops?Ecol. App1. 3:256-261. experiments.Catbirds were captured under permit COHEN,g. g. 1966. Anticoagulation,centrifugation PRT-761796(U.S. Fish and Wildlife Service).C. Shep- time and sample replicate number in the micro- pard and A. Lyles generously provided the Purple- hematocrit method for avian blood. Poult. Sci. headed Glossy-Starlings and commented on the 46:214-218. manuscript.L. Eguiarte,K. Brugger,D. Levey,A. Mas- DAHLQVIST,A. 1984. Assayof intestinal disacchari- ters, and D. O'Brien read and criticized drafts of the dases. Scand. J. Clin. Lab. Invest. 44:69-172. manuscript.D. W. White, K. G. Murray, and G. D. DENSLOW,J. S., D. J. LEVEY,T. C. MOERMOND,AND g. Schnell constructivelyreviewed the manuscript.Our C. WENTWORTH.1987. A syntheticdiet for fruit- work was funded by an NSF grant to C.M.R. (BSR- eating birds. Wilson Bull. 99:131-134. 9020909). GROSCOLAS, R., AND A. RODRIGUEZ. 1981. Glucose metabolismin fed and fastedEmperor Penguins LITERATURE CITED (Aptedonytesforsteri). Comp. Biochem.Physiol. A Comp. Physiol.70:191-198. ALPERS,D.H. 1987. Digestion and absorptionof car- GUDMAN-HOYER,E., P. A. KRASILNIKOFF,AND J. SKOV- bohydrates and proteins. Pages 1469-1486 in I•ERJ.1984. Sucrose-isomaltosemalabsorption. Physiologyof the gastrointestinaltract, vol. 2 (L. Pages35-62 in Advances in nutritional research, R. Johnson, Ed.). Raven Press,New York. vol. 6 (H. H. Draper, Ed.). Plenum Press,New AURICCHIO, S., A. DAHLQVIST, G. MORSET,AND A. PP,•- York. DER. 1963. Isomaltose intolerance causing de- HAZELWOOD, R.L. 1984. Pancreatic hormones, in- creasedability to utilize dietary starch.J. Pediatr. sulin/glucagon molar ratios, and somatostatinas 62:165-173. determinantsof avian carbohydratemetabolism. AURICCHIO,S., F. CICCIMARRA,L. MOAURO, F. REY, J. J. Exp. Zool. 232:647-652. Jos, AND J. REy. 1972. Intraluminal and mucosal HERNANDEZ,A., AND C. MARTINEZ DEL RIO. 1992. In- starchdigestion in congenitaldefficiency of intestinal disaccharidasesin five speciesof phyl- testinal sucrase and isomaltase activities. Pediatr. lostomoidbats. Comp. Biochem.Physiol. B Comp. Res. 6:832-840. Biochem. 10:105-111. BAKER,H. G., ANDI. BAKER.1983. Floral nectarsugar HOYSAK, D. J., AND P. J. WEATHERHEAD. 1991. Sam- constituentsin relation to pollinator type. Pages pling blood from birds: A technique and an as- 131-171 in Handbook of experimental pollina- sessment of its effects. Condor 93:746-752. tion ecology (C. E. Jonesand R. J. Little, Eds.). ISOKOSI,g., J. JUSSILA,AND S. SARNA. 1972. A simple Scientific and Academic Editions, New York. screeningmethod for lactosemalabsorption. Gas- BAKER,I., H. G. BAKER,AND S. A. HODGES. 1993. Pat- troenterology 62:28-32. terns in the sugarcomposition of nectar and fruit KRASILNIKOFF, P. A., E. GUDMAN-HoYER, AND I•I. H. juices taken by Microchiroptera and Megachi- MOLTKE. 1975. Diagnosticvalue of disacchari- roptera. Biotropica.In press. dase tolerance tests in children. Acta Paedtr. BEAL, F. E. L. 1897. Some common birds in their Scand. 64:693-702. relation to agriculture. U.S. Dep. Agric. Farmers MACKWORTH-PRAED,C. W., AND C. H. B. GRANT. 1955. Bull. 54. Birdsof easternand northern Africa. Longmans, BEECHER,W. J. 1978. Feeding adaptationsand evo- Green and Co., New York. January1994] SucroseIntolerance inBirds 177

MARSH, R. L., C. CAREY, AND W. R. DAWSON. 1984. SEMENZA,G., AND S. AURICCHIO. 1989. Small intes- Substrateconcentrations and turnover of plasma tinal disaccharidases.Pages 2975-2997 in The glucoseduring cold exposurein seasonallyac- metabolic basis of inherited disease, vol. 2 (C. R. climatized House Finches, Carpodacusmexicanus. Scriver, A. L. Beaudet, W. S. Sly, and D. Valle, J.Comp. Physiol.B Comp. Blochem.154:469-476. Eds.). McGraw-Hill, New York. MARTIN, A. C., H. C. ZIM, AND A. L. NELSON. 1951. SERLE,W., G. J. MOREL, AND W. HARTWIG. 1977. A Americanwildlife and plants:A guide to wildlife field guide to the birds of West Africa. William food habits. McGraw-Hill, New York. Collins Sons and Co., Toronto. MARTINEZDEL RIO, C. 1990. Dietary and phyloge- SIBLEY,C. G., AND J. E. AHLQUIST. 1984. The rela- netic correlates of intestinal sucrase and maltase tionships of the starlings (Sturnidae: Sturnini) activity in birds. Physiol. Zool. 63:987-1011. and the mockingbirds(Sturnidae: Mimini). Auk MARTINEZDEL RIO, C., ANDB. R. STEVEaXIS.1989. Phys- 101:230-243. iological constraint on feeding behavior: Intes- SIBLEY,C. G., AND J. E. AHLQUIST.1990. Phylogeny tinal membrane disaccharidasesof the starling. and classification of birds. Yale Univ. Press, New Science 243:794-796. Haven, Connecticut. MARTINEZ DEL RIO, C., B. R. STEVENS,D. DANEKE,AND SIBLEY,C. G., AND B. L. MONROE. 1990. Distribution P. T. ANDREADIS.1988. Physiological correlates and taxonomyof birds of the world. Yale Univ. of preference and aversion for sugars in three Press, New Haven, Connecticut. speciesof birds. Physiol. Zool. 61:222-229. SKADHAUGE,E. 1981. Osmoregulation in birds. MARTINEZDEL RIO, C., D. J. LEVEY,AND W. H. KARASOV. Springer-Verlag, Berlin. 1989. Physiological basis and ecological conse- STEVENS,C. E. 1988. Comparativephysiology of the quencesof sugarpreferences in CedarWaxwings. vertebrate digestive system. Cambridge Univ. Auk 106:64-71. Press,Cambridge. MARTLNEZ DEL RIO, C., H. G. BAKER, AND I. BAKER. SUTHERS,H. B., ANDD. D. SUTHERS.1990. Aging and 1992. Ecologicaland evolutionary implications sexingGray Catbirdsby external characteristics. of digestive processes:Bird preferencesand the N. Am. Bird Bander 15:45-52. sugar constituentsof floral nectar and fruit pulp. VONK, H. J., AND J. R. H. WESTERN.1984. Compar- Experientia 48:544-550. ative biochemistryof enzymatic digestion. Aca- MOSTELLER,F., ANDJ. W. TUKEY. 1977. Data analysis demic Press, London. and regression.Addison Wesley, New York. WEAST, R. C., AND S. M. SELBY. 1967. Handbook of NAGY, K. A. 1987. Field metabolic rate and food chemistryand physics,48th ed. ChemicalRubber requirement scaling in mammals and birds. Ecol. Co., Cleveland, Ohio. Monogr. 57:11-128. WHISTLER,R. L., AND J. R. DANIEL. 1985. Carbohy- PRINZINGER,R., A. PRESMAR,AND E. SCHLEUCER.1991. drates. Pages 69-138 in Food chemistry (O. R. Body temperature in birds. Comp. Biochem. Fennema, Ed.). Marcel Decker, Inc., New York. Physiol. A Comp. Biochem. 99:499-506. WHITE, D. W.,AND E. W. STILES. 1990. Co-occurrences PROP,J., AND T. VULINK. 1992. Digestionby Barnacle of food in stomachsand fecesof fruit-eating birds. Geesein the annualcycle: The interplaybetween Condor 92:291-303. retentiontime and foodquality. Funct. Ecol. 6:180- 189.