Mechanically fractionated flour isolated from green plantains (M. cavendishii var. nanica) as a tool to increase the dietary fiber and phytochemical bioactivity of layer and sponge cakes

Accep ed Manusc ip Mechanically ac iona ed lou isola ed om g een bananas (M. ca endishii a . nanica) as a ool o inc ease he die a y ibe and phy ochemical bioac i i y o laye and sponge cakes C is ina Segundo, Lau a Román, Manuel Gómez, Ma io M. Ma ínez PII: S0308-8146(16)31538-2 DOI: h p://dx.doi.o g/10.1016/j. oodchem.2016.09.143 Re e ence: FOCH 19923 To appea in: Food Chemis y Recei ed Da e: 4 July 2016 Re ised Da e: 21 Sep embe 2016 Accep ed Da e: 22 Sep embe 2016 Please ci e his a icle as: Segundo, C., Román, L., Gómez, M., Ma ínez, M.M., Mechanically ac iona ed lou isola ed om g een bananas (M. ca endishii a . nanica) as a ool o inc ease he die a y ibe and phy ochemical bioac i i y o laye and sponge cakes, Food Chemis y (2016), doi: h p://dx.doi.o g/10.1016/j. oodchem. 2016.09.143 This is a PDF ile o an unedi ed manusc ip ha has been accep ed o publica ion. As a se ice o ou cus ome s we a e p o iding his ea ly e sion o he manusc ip . The manusc ip will unde go copyedi ing, ypese ing, and e iew o he esul ing p oo be o e i is published in i s inal o m. Please no e ha du ing he p oduc ion p ocess e o s may be disco e ed which could a ec he con en , and all legal disclaime s ha apply o he jou nal pe ain. Mechanically ac iona ed lou isola ed om g een bananas (M. ca endishii a . nanica) as a ool o inc ease he die a y ibe and phy ochemical bioac i i y o laye and sponge cakes Running i le: Nu i ional enhancemen o cakes h ough ac iona ed banana lou s C is ina Segundo1,2, Lau a Román2, Manuel Gómez2, Ma io M. Ma ínez2,3* 1Facul ad de Ingenie ía, CIT Jujuy. Uni e sidad Nacional de Jujuy. CONICET. I alo Palanca, 4600 Jujuy, A gen ina 2Food Technology A ea. College o Ag icul u al Enginee ing. Uni e si y o Valladolid, 34004 Palencia, Spain. 3Whis le Cen e o Ca bohyd a e Resea ch, Depa men o Food Science, Pu due Uni e si y, Wes La aye e, IN 47907, U.S.A. *Co esponding au ho e-mail: ma 1269@pu due.edu: ma ioma [email protected] Abs ac This a icle desc ibes he e ec o mechanically ac iona ed lou s om g een bananas on he nu i ional, physical and senso y a ibu es o wo ypes o cakes (sponge and laye ). A plausible 30% eplacemen o banana lou in he o mula ion o laye cakes is demons a ed, inding only a small decline in he senso y pe cep ion. On he con a y, sponge cakes we e no iceable wo sened wi h he use o banana lou s (lowe speci ic olume, wo se senso y a ibu es and highe ha dness), which was minimized when using ine lou . Bo h laye and sponge cakes exhibi ed an enhancemen o he esis an s a ch and die a y ibe con en wi h he eplacemen o g een banana lou (up o a i e- old imp o emen in RS pe o mance). Mo eo e , sponge cakes yielded mo e polyphenols and an ioxidan capaci y wi h banana lou s, especially wi h he coa se ac ion. The e o e, esul s showed ha a mechanical ac iona ion allowed a easible nu i ional enhancemen o cakes wi h he use o banana lou s. Keywo ds: un ipe banana, die a y ibe , cake, pa icle size, an ioxidan capaci y, esis an s a ch 1. In oduc ion Die a y ibe (DF) ep esen s a b oad ange o ca bohyd a e compounds a ying in s uc u e ha scape diges ion and abso p ion wi hin he uppe gas oin es inal pa . High ibe die s may also be help ul o ecal bulking, dec eased ansi ime, educ ion o pos p andial glycemic esponse, main enance o no mal blood choles e ol le els, and a lowe isk o de eloping co ona y hea disease (Ve sp ee , e al., 2016). Despi e i s physiological bene i s, die s con ain a low amoun o DF, on a e age anging om 12-14g o die a y ibe pe day in Uni ed S a es and 18-20 g ams in Eu ope (Schmi z, Ma qua , and an de Kamp, 2007). Among die a y ibe s, esis an s a ch (RS) is he s a ch ac ion ha scape luminal diges ion. I encompasses i e di e en s a ch sub- ypes based on he mechanism by which hey esis diges ion. In pa icula , esis an s a ch ype 2 (RSII) is de ined as s a ch g anules ha a e inhe en ly esis an due o he composi ion and con o ma ion o he g anule. Fo mos sou ces o RSII, including po a o and banana, no mal cooking empe a u es, such as hose ound in baking, will end o dis up he s uc u e o he g anule, po en ially leading o gela iniza ion and inc easing i s diges ibili y (Bi ke and B own, 2007). This loss o die a y RS in wes e n die s con ibu es o wha has been e med he “ca bohyd a e gap”, which is he obse ed de ici in he amoun o e men able nu ien s compa ed wi h wha is equi ed by he in es inal mic o lo a o main ain diges i e heal h. In his scena io, die s ha e e se he cu en ends and inc ease die a y RS in ake a e impo an s a egies o ec i y his gap o imp o ing o e all heal h (Bi ke and B own, 2007). Banana is a gene al e m emb acing a numbe o species o hyb ids in he genus Musa o he amily Musaceae. In pa icula , ca endish subg oup banana cul i a s (M. ca endishii) a e he mains ays o he expo ades (Zhang, Whis le , BeMille , and Hamake , 2005).. Wi h a global p oduc ion o 106 million ons in 2013 (FAO, 2015), abou one- i h o all bananas ha es ed become culls. When banana bunches a i e a cen al collec ion s a ions, bananas oo small o shipping a e emo ed, along wi h hose ha ha e damaged o spoiled a eas ha could cause mic obial con amina ion o he bunch. Rejec ed bananas a e no mally disposed o imp ope ly. The e o e, a success ul indus ial use o he culled bananas, such as he p oduc ion o a low-cos banana lou ing edien , would alle ia e he p oblem. The igge o he indus ial use o banana lou could be based on i s nu i ional p ope ies. In pa icula , banana lou con ains an impo an ac ion o die a y ibe [up o 15% in some a ie ies, (da Mo a, Lajolo, Ciacco, and Co denunsi, 2000)], consis ing mainly o pec in (soluble ac ion) and cellulose, lignin and hemicellulose (insoluble ac ion) (Thebaudin, Le eb e, Ha ing on, and Bou geois, 1997). Among he die a y ibe , na i e aw banana s a ch is also known o be esis an o he a ack o diges i e enzymes, wi h in i o esul s showing ha 75- 84% o he s a ch g anules inges ed eached he e minal ileum (Englys and Cummings, 1986; Faisan , e al., 1995). Al hough i was ound ha he esis ance was la gely o e come by cooking o gela inize he s a ch, o he s udies showed ha he ‘easily hyd olysable s a ch’ could be lowe a e cooking compa ed o o he s a ch sou ces (Zhang, and Hamake , 2012). Ano he impo an nu i ional bene i o bananas is hei an ioxidan capaci y a ibu ed o he ole in he sca enging o ee adicals o phenolic compounds, which con e exis ing ee adicals in o less ha m ul molecules and epai ing oxida i e damage (Du, Li, Ma, and Liang, 2009). E en hough he an ioxidan p ope ies ha e been especially a ibu ed o banana peels (Kondo, Ki iko n, and Kanlayana a , 2005), accumula ing e idence has e ealed ha banana pulp also con ains a ious an ioxidan s, such as phenolic compounds, comp ising ca echin, epica echin, lignin and annins and an hocyanins (Sulaiman, e al., 2011). All hese nu i ional bene i s ha e made banana lou o be explo ed as a unc ional ing edien in a ious p oduc s such as slowly diges ible cookies (Apa icio-Saguilan, e al. 2007), high- ibe b ead (Ho, Aziz, and Azaha i, 2013; Jua ez-Ga cia, Agama-Ace edo, Sayago-Aye di, Rod iguez-Amb iz, and Bello-Pe ez, 2006), pas a wi h highe an ioxidan capaci y (O ando- Ma inez, Sayago-Aye di, Agama-Ace edo, Goñi, and Bello-Pe ez, 2009, Almanza-Beni ez, Oso io-Díaz, Mendez-Mon eal o, Islas-He nandez, and Bello-Pe ez, 2015), noodles (Choo, and Aziz, 2010), o illas (Apa icio-Sanguilan, e al., 2013) and snacks (Wang, Zhang, and Mujumda , 2012). Howe e , o he bes o ou knowledge, he use o banana lou as an ing edien in baked emulsions, such as cakes, has been no conside ed so a . Banana lou coun s wi h 61-76.5% s a ch (da Mo a, e al., 2000), which will de e mine some o he lou unc ionali y. Howe e , he in insic p ope ies o lou pa icles, such as pa icle size, will s ongly in luence hei emulsi ying and hyd a ion p ope ies and he e o e a ec ing he quali y o he esul an cakes. In pa icula , ice lou s wi h coa se pa icle size did no allow he co ec emulsion o ma ion by hal ing ai inco po a ion in o m o ine and homogenous bubbles, wo sening cake olume and ex u e (de la He a, Ma ínez, Olie e, and Gomez, 2013b). Howe e , he op imum lou pa icle size dis ibu ion o cake-making depends, in u n, on he ype o cake. In pa icula , Dhen, e al., (2016) epo ed a nega i e in luence o cou se soy lou on he quali y o sponge cake, whe eas no signi ican di e ences we e obse ed o laye cakes. The e o e, ou hypo hesis was ha he use o lou isola ed om g een bananas wi h an op imum pa icle size could: 1) imp o e he nu i ional p ope ies o cakes while main aining i s physical and senso y a ibu es, and, 2) minimize he was e o banana culls. The objec i e was o s udy he e ec o a whea lou eplacemen (15 and 30%) by g een banana lou wi h h ee di e en pa icle sizes [80µm ( ine), 80-156µm (medium) and 156-200 µm (coa se)] on he nu i ional (die a y ibe , esis an s a ch, phenolic compounds, an ioxidan capaci y), physical (speci ic olume, ex u e and colo ) and senso y p ope ies o laye and sponge cakes. To do his, banana lou was ob ained om g een (un ipe) bananas h ough a low- cos p ocedu e and subsequen ly sie ed. Nu i ional and physicochemical p ope ies o lou s as well he heological p ope ies o ba e s we e also analyzed in o de o enhance he discussion abou he di e en phenomena in ol ed. 2. Ma e ials and me hods 2.1 Ma e ials Whea lou (10.3 g/100 g mois u e; 8.88 g/100 g p o ein; 84.23 µm mean pa icle size) was supplied by Ha ine a la Cas ellana S.A (Medina de Campo, Valladolid, Spain). Suga , sun lowe oil, whole liquid milk, powde ed milk and liquid pas eu ized egg we e pu chased om he local ma ke . Baking powde 2x1 and he emulsi ie Supe Mixo T500 we e p o ided by Pu a os (Ge ona, Spain). Bananas (M. ca endishii a . nanica) we e pu chased a ipening s a e 2 (yellow wi h black spo s) acco ding o s anda d colo cha , h p:// ipening- ui .com/banana). 2.2 Me hods 2.2.1 Isola ion o banana lou om g een bananas and lou cha ac e iza ion Bananas we e peeled and g ound in a Liliana AM523 juice (Che y, A gen ina) a 1 speed o 3 min. The pas e ob ained was pou ed in a ay and d ied in an o en wi h o ced con ec i e low a 60°C o 24 hou s. D ied pas e was g ound in a Numak FW 100 labo a o y mill (Ojal o, A gen ina) and subsequen ly sie ed in a Bühle MLI 300B sc eening machine (Uzwil, Swi ze land) o 15 min, ob aining a ine (<80µm), medium (80-156µm) and coa se ac ions (156-200µm). The di e en ac ions we e sealed in plas ic bags and s o ed. 2.2.2 P oxima e analysis and cha ac e iza ion o lou s Banana lou was cha ac e ized ollowing AACC (2016) me hods o mois u e con en (44.15.02), p o ein (46.11.02), lipids (30.25.01), die a y ibe (32.05.01), ash (08.12.01) and esis an s a ch (32-40.01). Ca bohyd a e con en o lou s was ob ained by weigh di e ence wi h he o he lou componen s. The amylose con en was de e mined by he Megazyme amylose assay ki (Megazyme, B ay, I eland) using he Concana alin A p ecipi a ion p ocedu e. I is no ewo hy ha he esis an s a ch con ained in g een banana is comp ised mainly by s a ch g anules ha a e inhe en ly esis an due o he composi ion and con o ma ion o he g anule (RSII). Howe e , his ype o RS is dis up ed by high empe a u es, such as hose in he ini ial s ep o o al ibe analysis, po en ially leading o gela iniza ion (Bi ke and B own, 2007). The e o e, RSII was no included in he ibe analysis (me hod 32.05.01, AACC, 2016). All analyses we e un in duplica e. The cha ac e is ics o he di e en banana lou ac ions a e shown in Table 1. Flou pho omic og aphs we e aken wi h a Quan a 200FEI (Hillsbo o, O egon, USA) ESEM (Fig. 1). Pho omic og aphs we e aken in beam decele a ion mode (BDM) a 1.5 keV in high acuum mode wi h a backsca e ed elec on de ec o (BSED). 2.2.3 Cake making Laye and sponge cake we e made as ollows. Laye cakes consis ed o 350 g lou , 315 g suc ose, 210 g whole milk, 175 g liquid pas eu ized egg, 105 g sun lowe oil and 10.5 g baking powde . All ing edien s we e mixed in he Ki chen Aid P o essional mixe KPM5 (S . Joseph, Michigan, USA) o 1 min a speed 4 and 9 min a speed 6. Cake ba e (185 g) was placed in o disposable oil-coa ed aluminum pans (109 x 159 x38 mm) and he cakes we e baked in an elec ic o en o 25 min a 185 °C. Sponge cake o mula consis ed on 245 g lou , 240.5 g suga , 344 g liquid pas eu ized egg, 55 g wa e , 14 g emulsi ie , and 25 g powde ed milk. A c eaming-mixing p ocedu e was used. All ing edien s, excep o he lou and milk we e mixed o 2 min a speed 6 using a Ki chen-Aid P o essional mixe (Ki chen Aid, S . Joseph, Michigan). A e he addi ion o he milk and lou , he mixing p ocess was con inued o 3 min a speed 8. Cake ba e (125 g) was placed in o oil- coa ed aluminum pans (109x 159 x 38 mm) and baked as desc ibed abo e. A e baking, cakes we e emo ed om he pans and le o cool o 60 min be o e being placed in coded plas ic bags ha we e sealed o p e en d ying. Fo each ype o cake (laye o sponge), a 15 and a 30% o whea lou eplacemen by ine, medium and coa se banana lou was s udied, which oge he wi h he con ol (100% whea lou ) ga e 7 o mula ions. All cakes we e made in duplica e. 2.2.4 Densi y and heological p ope ies o ba e s Ba e densi y was de e mined wi h a s anda d con aine (100 cc) o known weigh illed wi h ba e . Be o e conduc ing any heological measu emen , samples we e allowed o es in he measu emen posi ion o 5 min as equilib a ion ime, i.e. he necessa y ime o allow he s esses induced du ing sample loading o elax. The exposed edges o he samples we e always co e ed wi h aseline oil (Pan eac Química S.A., Cas ella del Valles, Spain) o a oid sample d ying du ing measu emen s. Rheological p ope ies we e eco ded using a heome e (Haake RheoS ess 1, The mo Fische Scien i ic, Sche e e, Ge many). All heological es s we e un in duplica e. 2.2.4.1. Flow p ope ies. (Lakshmina ayan, Ra hinam, and K ishnaRau, 2006). The di e en end epo ed o each ype o ba e (sponge and laye ) sugges s ha he e ec s o banana lou depend on he ype o emulsion. In sponge ba e s, he emulsion is o med by a g ea e numbe o ai bubbles o med du ing mixing, which a e mo e closely packed han in laye cakes, as a esul o he emulsi ie ac ion (Gómez, Ruiz, and Olie e, 2011). On he o he hand, he baking powde in laye cakes, esponsible o he ai bubbles g ow h, does no ac un il a ce ain empe a u e is eached in he o en. In his way, in sponge cake emulsion, coa se pa icles would di icul he emulsi ie ac ion by hal ing bubble s abiliza ion (Dhen, e al., 2016) and p omo ing hei up u e du ing he shea s ess p oduced du ing he o a ional heological es . This hypo hesis is endo sed in u n by he signi ican loss o pseudoplas ic beha io (highe low beha io index, n) obse ed in sponge cakes subs i u ed by he medium and coa se ac ions o banana lou , which was less e iden in laye cakes. A lowe pseudoplas ic beha io en ails a less complex s uc u e o hese ba e s, leading o an emulsion beha io mo e simila o wa e , which is, in ac , in ag eemen wi h he heo y o he ai bubbles escaping om he ba e du ing he o a ional low beha io es . In o de o be e cha ac e ize he s uc u e o ba e s, he small ampli ude oscilla o y es was expec ed o p o ide mo e accu a e in o ma ion. The in luence o he addi ion o sie ed banana lou s on he iscoelas ici y is shown in Table 2. In he a ailable equency window, he so- called pla eau elaxa ion zone was obse ed in all he ba e s (da a no shown). This egion is cha ac e ized by he ac ha he s o age modulus (G’) is highe han he loss modulus (G’’), bo h moduli depending on equency bu ollowing a di e en pa e n (Ma inez, Sanz, and Gomez, 2015). This egion is cha ac e is ic o he occu ence o physical en anglemen s in polyme ic ma e ials (Fe y, 1980). In sponge cakes, a signi ican inc ease o G’ was obse ed wi h a 15 and 30% eplacemen o medium banana ac ions, whe eas in laye cakes, he coa ses ac ion led o he mos no iceable inc ease o bo h iscoelas ic moduli. This would sugges ha ba e s con aining medium and coa se banana lou s ha e he same heological uni s han he es bu wi h an enhanced con ibu ion o inc ease he ba e consis ency (Ma inez, e al., 2015). Howe e , a 30% subs i u ion by he medium ac ion (sponge cake) and he coa se ac ion (laye cake) o banana lou had a signi ican ly lowe loss angen ( anδ), indica ing di e en iscoelas ici y and he e o e he p esence o new heological uni s. This can be a ibu ed o he ac ha iscoelas ic p ope ies a e also dependen on he concen a ion o ai apped du ing mixing (Kalinga, and Mish a, 2009). Acco ding o he lou composi ion (Table 1), i seems ha ibe con en , ha ing a much highe wa e abso p ion capaci y han in ac s a ch, could diminish he ee wa e con en enhancing he s uc u e and inc easing bo h dynamic moduli. The esul would be a masking o he e ec o he in insic highe su ace o olume a io o ine lou s. The ac ha hese esul s ollow a di e en end han hose o K in sponge cakes would con i m ha du ing he o a ional es he shea s ess was enough o p omo e he ai bubbles b eakage, while i did no happen du ing he oscilla ion sweeps. On he o he hand, he pa allel inc ease be ween he iscoelas ic moduli and K would sugges ha in laye emulsion, no bubble scape was p oduced. 3.3 Cake cha ac e is ics In o de o assess he e ec o he inco po a ion o he di e en ac ions o banana lou on he physical p ope ies o cakes, speci ic olume and c umb ex u e we e analyzed (Table 3). The e ec s o he di e en banana lou ac ion we e condi ioned o he ype o cake. The speci ic olume o laye cakes was ba ely modi ied wi h he inclusion o banana lou (less han 8% o inc ease/dec ease compa ed o con ol), app ecia ing a sligh ly imp o e wi h ine and medium ac ions. Con e sely, he inco po a ion o g een banana displayed a clea nega i e e ec on he speci ic olume o he sponge cakes, which was educed g ea ly when inc easing he pa icle size o banana lou and he le el o eplacemen . In ac ual ac , he only cake which was no signi ican ly di e en om he con ol was he cake made wi h 15% o ine banana lou , whe eas wi h 30% coa se banana lou eplacemen he educ ion in he speci ic olume eached 36%. This esul s a e in ag eemen wi h he consis ency and low beha io index o sponge ba e s, indica ing ha highe consis ency (highe K) and pseudoplas ici y (lowe n) b ing abou high olume cakes. Howe e , his co ela ion was no clea o laye cakes, whe e bubbles a e c ea ed du ing baking. In his way, changes in olume may no be solely ela ed o he ai inco po a ed du ing he mixing, bu o di e ences in he ai e en ion and expansion du ing baking p ocess. I is known ha bubble size is c ucial o ba e s abili y in such a way ha , big a e age bubble size is ela ed wi h low s abili y, especially du ing he i s s ages o baking, when he ba e iscosi y is educed (S au e , 1990). The in e nal s uc u e o sponge cakes consis s o a g ea e numbe o bubbles wi h smalle a e age size han laye cakes (Gómez, e al., 2011). This s uc u e could be mo e sensi i e o he p esence o coa se lou pa icles, which could impac mo e nega i ely o he olume o sponge cakes han he olume o laye cakes (de la He a, e al., 2013b; Dhen, e al., 2016). C umb ha dness o sponge cakes was highly co ela ed o he speci ic olume o cakes, obse ing a nega i e co ela ion be ween speci ic olume and c umb ha dness ( =-0.71), which ag ees wi h ha obse ed in o he s udies (Gómez, Ruiz-Pa ís, Olie e, and Pando, 2010; de la He a e al., 2013b). These s udies ha e al eady p o en ha his co ela ion was mo e signi ican in sponge han in laye cakes. In ac , in ou s udy, he dec ease in c umb ha dness o laye cakes was no co ela ed o he speci ic olume and he e o e he phenomena canno be asc ibed o changes in olume. We hypo hesize ha i should be ela ed o s uc u al changes in he c umb, possibly in luenced by he di e ences be ween he s a ches om banana and whea in e ms o gela iniza ion and e og ada ion beha io s. In his way, despi e signi ican di e ences we e ha dly seen in he sp inginess o laye cakes, a decline in he cohesi eness was ound o he cakes made wi h 30% o banana lou , which may be ela ed o he di e ences be ween whea and banana s a ch. A simila end was ob ained o sponge cakes, bu in his case, bo h a educ ion in sp inginess and cohesi eness was app ecia ed, being his e ec clea e as he le el o eplacemen was inc eased, especially o he coa ses pa icle sizes. 3.4 Senso y e alua ion The senso y analysis (Table 4) was pe o med in laye cakes wi h 30% banana lou eplacemen and in sponge cakes wi h 15% eplacemen , since sponge cakes wi h 30% eplacemen showed a low speci ic olume. The inclusion o banana lou in laye cakes led o a sligh wo sening in he accep abili y o he cakes, bu no e ec was ound o he pa icle size, which in gene al, is in acco dance wi h he esul s obse ed o he physical cha ac e is ics o cakes. The mos impo an di e ences among he pa ame e s s udied we e no iced in appea ance, possibly due o colou di e ences (supplemen a y ma e ial), and odo . Changes in he colo o he c us should be explained by he p esence o highe suga s in banana lou s, such as glucose, uc ose and suc ose (Emaga, And ianai o, Wa hele , Tchango, and Paquo , 2007). These suga s a e p ecu so s o Mailla d eac ions and ca ameliza ion because o he high empe a u es eached on he ex e nal pa o he cake du ing baking (Pu lis, 2010). As o he inne pa o he cake, i.e., c umb, which do no exceed 100°C du ing baking, Mailla d and ca ameliza ion eac ions do no ake place, and colo di e ences can be a ibu ed o di e ences in he colo o he lou . In his case banana lou is less whi ish han whea lou , wi h sligh di e ences in he pa icle size (see supplemen a y ma e ial). In gene al, di e ences o o e all cake accep abili y o cakes made wi h banana lou we e no g ea e han 0.7 compa ed o he con ol. In he case o he sponge cakes, despi e he ac ha he le el o eplacemen chosen was lowe , a lowe gene al accep ance o cakes was obse ed wi h banana lou eplacemen (excep o cakes made wi h he ine ac ion), being di e ences mo e no iceable when pa icle size was inc eased. These esul s a e in ag eemen wi h he physical cha ac e is ics o cakes epo ed in he p e ious sec ion. Acco ding o he esul s, laye and sponge cakes we e ob ained wi h a low decline in hei accep abili y wi h a 30% and 15% banana lou eplacemen , espec i ely. Fu he mo e, he e ec o lou pa icle size depended, in u n, on he ype o cake. In pa icula , laye cakes we e no in luenced by he pa icle size whe eas sponge cakes wi h no signi ican di e ences wi h he con ol we e only ob ained wi h banana lou wi h ine pa icle size.3.5 Resis an s a ch, die a y ibe , phenolic compounds and an ioxidan capaci y Laye and sponge cakes made wi h he di e en ac ions o g een banana lou we e analyzed o esis an s a ch, o al die a y ibe , phenolic compounds and an ioxidan capaci y (Fig. 2). Bo h laye and sponge cakes exhibi ed an enhancemen o he die a y esis an con en wi h he eplacemen o g een banana lou , especially wi h he highe le el o subs i u ion (30%). In pa icula , he use o g een banana lou s yielded up o a i e- old imp o emen in RS pe o mance (Fig. 2a). Al hough i has been epo ed ha he esis ance o banana s a ch in hei na i e s a e o he luminal diges ion is o e come by hyd o he mal p ocessing, ou esul s show ha he e is an impo an ac ion ha pe sis s in he inal cake. This would indica e ha ei he he in eg i y o some g anules could esis he high empe a u es a ela i ely high mois u e le els du ing baking o ha a e cooling some leached amylose chains would e og ade c ea ing esis an s a ch ype 3 (RSIII). Howe e , we belie e ha he o me is a mo e plausible mechanism, since cake sys ems a e also composed by o he ing edien s such as suga s, ha could delay s a ch gela iniza ion and he e o e keeping he g anula in eg i y. The inhe en esis ance o he in eg al g anules is in u n ein o ced by i s B- ype c ys allini y (en ailing lack o supe icial po es) and he e o e wi hou channels o he access o panc ea ic α-amylase du ing in aluminal diges ion. Ou esul s a e in ag eemen wi h o he wo ks a e p ocessing banana hyd o he mally (Zhang, e al., 2012). Resul s also showed ha 30% eplacemen wi h ine lou s ga e ise o he highes RS, which seems logical ega ding he highe RS con en o hese lou s in hei na i e s a e (Table 1). Explana ions o his occu ence a e gi en in sec ion 3.1. The enhancemen in he RS was also complemen ed wi h he inc ease o o he sou ces o die a y ibe h ough he subs i u ion o g een banana lou (Fig. 2a). In pa icula , hese ibe s would encompass pec in, cellulose, lignin and hemicellulose, ypical om banana lou s (Thebaudin, e al., 1997). In his case, a highe amoun o die a y ib e was ound o he coa ses ac ions o g een banana lou , which was also expec ed acco ding o he ibe con en o he ini ial lou s (Table 1). Explana ions o his occu ence a e also gi en in sec ion 3.1. In sponge cakes, a signi ican inc ease o he polyphenol con en was obse ed o he medium and coa se ac ions o banana lou s (Fig. 2b), which is in acco dance wi h polyphenol con en o he ini ial lou s. As o laye cakes, e en hough a highe phenolic con en can be obse ed, he e ec was no as no iceable as in sponge cakes. The easons o his di e en beha io mus be app oached om he ac ha high empe a u es (abo e 50°C) can des oy phenolic compounds (Azizah, Ruslawa i, and Swee-Tee, 1999). As we men ioned, he in e nal s uc u e o he di e en ype o cake a ies (Gómez, e al., 2011). The g ea e numbe o bubbles wi h smalle a e age size in sponge cakes b ings abou ba e s wi h highe consis ency and solid-like beha io han laye ba e s (see Table 3), which can lead o mode a ing he mass and hea ans e s and he e o e diminishing he loss o phenolic compounds du ing baking. In o he wo ds, he sho baking ime and he glu en-s a ch-non s a ch polysaccha ides ma ix in he cake subs i u ed wi h banana lou s could p o ec he polyphenols agains excessi e des uc ion (Almanza-Beni ez, e al., 2015). This occu ence would happen wi h less s eng h in laye cakes, in which he loss o phenolic compounds om he aw banana lou s would be highe as a consequence o he absence o en i ies ha p o ec phenolic compounds om des uc ion du ing baking. The inco po a ion o g een banana lou s b ough abou sponge and laye cakes wi h lowe alues o IC50, i.e., lowe e ec i e concen a ion a which he an ioxidan ac i i y was 50% and he e o e indica ing highe phy ochemical ac i i y. In pa icula , cakes made wi h he coa ses ac ions o banana lou s had he highes an ioxidan capaci y. This ollowed he same end as he polyphenols con en . F ui s a e usually ich in polyphenols such as an hocyanins and p oan hocyanindins which can inc ease he sca enging ac i i y o he s able DPPH adical model, as a consequence o he hyd ogen dona ing abili y (Von Gadow, Joube , and Hannsman, 1997). In addi ion, hose phenolic compounds could be associa ed wi h he highe p esence o die a y ibe , since i was epo ed ha die a y ibe e ains an ioxidan capaci y. Du ing e men a ion o die a y ibe in he colon, he phenolic compounds wi h an ioxidan capaci y can exe i s physiological ac i i y (Sau a-Calix o, 1998). The signi ican an ioxidan capaci y o cakes, and in pa icula o sponge cakes subs i u ed wi h g een banana lou s, is an addi ional nu aceu ical cha ac e is ic o ha o he high RS and die a y ibe con en . In his sense, consump ion o cakes subs i u ed wi h g een banana lou s could gi e mo e bene icial e ec s o people's heal h. 4. Conclusion In his s udy, we showed ha lou isola ed om g een bananas can be used o imp o e he nu i ional p ope ies o laye cakes wi hou nega i ely a ec ing hei physical p ope ies, esul ing in only a small decline in he senso y pe cep ion. On he o he hand, he physical and senso y a ibu es o sponge cakes we e no iceably wo sened wi h he use o banana lou s, bu his e ec was minimized when using ine lou . Inc eases o RS and die a y ibe (o he han RS) con en we e a ained in bo h ype o cakes wi h g een banana lou eplacemen . Fu he mo e, he inco po a ion o g een banana lou s also enhanced he con en o phenolic compounds and he an ioxidan capaci y o cakes, especially o sponge cakes, whe e a 40% inc ease o an ioxidan capaci y was achie ed. As o he pa icle size, coa se lou s yielded cakes wi h highe con en o die a y ibe , phenolic compounds and an ioxidan capaci y han hose made wi h he ine coun e pa s. Howe e , hey also led o a signi ican decline in he olume and ex u al p ope ies o sponge cakes, e en hough no signi ican di e ences in he senso y pe cep ion we e ound be ween he di e en pa icle sizes. On he o he hand, ine lou s yielded cakes wi h signi ican ly highe esis an s a ch con en (up o a i e- old imp o emen ), highe olume and lowe ha dness han he cou se coun e pa s. Thus, banana ine lou s would be mo e sui able in gene al. Howe e , he manu ac u e should also conside ha he op imum pa icle size o banana lou o cake-making is dependen on he nu i ional a ge as well as he ype o cake. 5. Acknowledgemen s The au ho s acknowledge he inancial suppo o he Spanish Minis y o Economy and Compe i i eness (P ojec AGL2014-52928-C2-2-R) and he Eu opean Regional De elopmen Fund (FEDER). Re e ences AACC. (2016). App o ed me hods o he Ame ican Associa ion o Ce eal Chemis s, 44- 15.02 (mois u e con en ), 46-11.02 (p o ein con en ), 08-12.01 (ash con en ), 30-25.01 (lipid con en ), 32-05.01 ( o al ibe con en ) and 32.40.01 ( esis an s a ch) (11 h ed.). S . Paul, Minneso a: Ame ican Associa ion o Ce eal Chemis s. Almanza-Beni ez, S., Oso io-Díaz, P., Mendez-Mon eal o, G., Islas-He nandez, J. J., & Bello-Pe ez, L. A. (2015). Addi ion o acid- ea ed un ipe plan ain lou modi ied he s a ch diges ibili y, indiges ible ca bohyd a e con en and an ioxidan capaci y o semolina spaghe i. LWT-Food Science and Technology, 62, 1127-1133. Apa icio-Saguilan A., Sayago-Aye di S. G., Va gas-To es A., To a J., Ascencio-O e o T. E., & Bello-Pe ez L. A. (2007). Slowly diges ible cookies p epa ed om esis an s a ch- ich lin ne ized banana s a ch. Jou nal o Food Composi ion and Analysis, 20, 175-181. Azizah, A. H., Ruslawa i, N. M., & Swee-Tee, T. (1999). Ex ac ion and cha ac e iza ion o an ioxidan om cocoa by-p oduc s. Food Chemis y, 64, 199-202. Bi ke , A. M., & B own, I. L. (2007). Resis an s a ch and heal h. In B. R. Hamake (Ed.), Technology o unc ional ce eal p oduc s (pp. 63-85). Boca Ra on, USA: CRC P ess. Choo, C. L., & Aziz, N. A. A. (2010). E ec s o banana lou and be a-glucan on he nu i ional and senso y e alua ion o noodles. Food Chemis y, 119, 34-40. da Mo a, R. V., Lajolo, F. M., Ciacco, C., & Co denunsi, B. R. (2000). Composi ion and unc ional p ope ies o banana lou om di e en a ie ies. S ä ke-S a ch, 52, 63-68. de la He a, E., Gomez, M., & Rosell, C. M. (2013a). Pa icle size dis ibu ion o ice lou a ec ing he s a ch enzyma ic hyd olysis and hyd a ion p ope ies. Ca bohyd a e Polyme s, 98, 421-427. Table 2. Densi y and heological p ope ies o ba e wi h 15 and 30% banana lou eplacemen Sponge ba e Laye ba e ρ G’(Pa) G’’(Pa) anδ K(Pa·s n ) n ρ G’(Pa) G’’(Pa) anδ K(Pa·s n ) Con ol 0 . 37 0 c 1080 a 327 ab 0 . 31 c 33 . 13 b 0 . 57 b 1. 10 e 134 a 111 a 0 .82 bc 20 . 31 ab Fine 15 0 . 355 a b 1195 ab 337 abc 0 . 29 bc 36 . 93 c 0 . 55 a 1.08 d 97 a 81 a 0 . 83 c 22 . 6 0 bc Medium 15 0 . 35 0 a 1430 b 409 c 0 . 29 bc 30 . 85 b 0 . 60 c 1.05 c 161 a 138 a 0 .85 c 18 . 67 a Coa se 15 0 . 36 0 b 1110 a 318 ab 0 . 29 bc 26 . 49 a 0 . 63 d 1.0 8 d 148 a 125 a 0 .84 c 17 . 21 a Fine 30 0 . 36 0 b 1085 a 310 a 0 . 29 bc 30 . 78 b 0 . 59 bc 1.0 6 c 75 a 64 a 0 .85 c 18 . 77 a Medium 30 0 . 36 0 b 1455 b 394 bc 0 . 27 b 27 . 38 a 0 . 64 de 1.0 1 b 453 b 346 b 0 . 77 b 24 . 53 cd Coa se 30 0 . 346 a 1305 ab 318 ab 0 .24 a 25 . 61 a 0 .65 e 0.99 a 3220 c 593 c 0.18 a 27 . 34 d Di e en le e s in he same column mean signi ican ly di e en (p<0.05) ρ, ba e densi y; G’, s o age modulus; G’’, loss modulus; anδ, loss angen ; K, consis ency coe icien ; n, low beha io index Table 3: Physical cha ac e is ics o sponge and laye cakes subs i u ed wi h banana lou o di e en pa icle sizes Sponge cake Laye cake Speci ic olume (cm3/g) Ha dness (N) Sp inginess Cohesi eness Speci ic olume (cm3/g) Ha dness (N) Sp inginess Cohesi eness Con ol 4.08 d 3.08 a 0.92 d 0.67 e 2.41 b 6 . 0 d 0.89 b Fine 15 3.94 d 3 . 38 a 0.88 bc 0.57 d 2.34 a 6 . 18 d 0.87 ab Medium 15 3.16 c 3.18 a 0.90 cd 0.57 d 2.6 d 4.36 a 0.88 b Coa se 15 3.02 bc 4.32 b 0.89 bc 0.51 cd 2.41 ab 4.90 b 0.86 ab Fine 30 3 . 34 c 4.53 b 0.89 bc 0.49 bc 2.60 d 5.15 bc 0.88 b 0.475 Medium 30 2.73 ab 6.93 c 0.87 ab 0.42 a 2.42 b 5.40 c 0.83 a Coa se 30 2.61 a 4.815 b 0.85 a 0.44 ab 2.5 c 4.19 a 0.85 ab Di e en le e s in he same column mean signi ican ly di e en (p<0.05) Table 4. Senso y p o ile o sponge and laye cakes subs i u ed wi h banana lou o di e en pa icle sizes Sponge cakes (15%) Laye cakes (30%) Appea ance Od o Tex u e Tas e Pe sis ence O e all accep ab ili y Appea ance Od o Tex u e Tas e Pe sis ence O e all accep ab ili y Con 7.08 c 6.2 6.43 a 6.6 6.30 a 6.78 b 7.54 b 6.5 6.32 b 6.6 6.37 a 6.93 b Fine 6.67 bc 5.8 6.29 a 6.4 6.19 ab 6.52 ab 5.97 a 5.6 5.59 a 6.2 6.25 a 6.41 a Medi 5.94 a 5.7 6.11 a 6.3 6.00 ab 6.24 a 6.03 a 5.5 5.80 a 6.0 6.01 a 6.23 a Coa 6.14 ab 5.7 5.97 a 6.1 5.75 a 6.11 a 5.94 a 5.3 5.80 a 6.3 6.08 a 6.38 a Di e en le e s in he same column mean signi ican ly di e en (p<0.05) Highligh s The coa se ac ions o banana lou yielded mo e ibe and phenolic compounds 30% eplacemen o banana lou in laye cakes is p o ed wi hou a quali y wo sening The pa icle size o banana lou s s ongly in luenced he quali y o sponge cakes Banana lou s inc eased he con en o esis an s a ch and die a y ibe o cakes Sponge made wi h banana lou s yielded mo e polyphenols and an ioxidan capaci y