Food Research International 81 (2016) 203–209

Contents lists available at ScienceDirect

Food Research International

journal homepage: www.elsevier.com/locate/foodres

Identification of carbohydrate parameters in commercial unripe

banana flour
Fabiana A. Hoffmann Sardá a,b, Fernanda N.R. de Lima a, Nathalie T.T. Lopes a, Aline de O. Santos a,b,
Eric de C. Tobaruela a,b, Edna T.M. Kato c, Elizabete W. Menezes a,b,⁎
a
Department of Food and Experimental Nutrition, Faculty of Pharmaceutical Sciences, University of São Paulo (USP), São Paulo, SP, Brazil
b
Food and Nutrition Research Center (NAPAN) — USP, Food Research Center (FoRC/CEPID/FAPESP), São Paulo, SP, Brazil
c
Department of Pharmacy, Faculty of Pharmaceutical Sciences, University of São Paulo (USP), São Paulo, SP, Brazil

a r t i c l e i n f o s u m m a r y

Article history: Unripe banana flour (UBF), which is rich in resistant starch (RS), has shown several positive physiological effects
Received 11 February 2015 in clinical trials. Although such observations encourage the emergence of UBF in the food market, specific identity
Received in revised form 6 November 2015 or quality standards for the product are still lacking. This work aimed to assess and propose characterization pa-
Accepted 15 November 2015
rameters for commercially available UBF. The results showed that three of the brands examined presented a RS
Available online 18 November 2015
content higher than 40%, whereas nine showed a lower content, with two having less than 10% RS and over 80%
Keywords:
total starch, which was fully identified as cereal starch by light microscopy (LM). The presence of banana peel in
Light microscopy the flour was correlated with the lipid (r = 0.870), ash (r = 0.812), protein (r = 0.704) and total starch
Quality standard (r = −0.761) contents. According to principal components analysis (PCA) and LM identification, the main pa-
Resistant starch rameters for the characterization of commercial UBFs are the contents of RS, dietary fiber, lipid and ash. The
large variability in RS content (4 to 62%) found in commercial UBFs is one reason why consumers would benefit
from additional labeling information, such as the inclusion of the RS and soluble sugar (SS) contents, the unripe
banana cultivar used, and indications about use of the peel. Moreover, adulterations could be verified by food in-
spection agencies via LM, which can be used as a tool to identify the type and state of the starch present.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction to 12 weeks) (Johnston, Thomas, Bell, Frost, & Robertson, 2010; Maki
et al., 2012; Robertson et al., 2012).
Functional foods and ingredients with a high content of unavailable Global banana production was approximately 102 million tons in
carbohydrates, such as resistant starch (RS), have shown promising po- 2012, and Brazil is the world's fifth largest producer, with an output
tential for reducing the incidence of non-communicable diseases of 6.9 million tons per year (FAO, 2014). However, of the total
(NCDs) due to the reduced speed of digestion and absorption of these amount produced in the country, approximately 30% is lost during
carbohydrates (Fuentes-Zaragoza, Riquelm-Navarrete, Sánchez-Zapata, the post-harvest phase (Almeida, 2012). Thus, the development
& Pérez-Álvarez, 2010; Birt et al., 2013). Over the years, interest and and utilization of any technology or process that enhances its use
knowledge about the biological utilization of RS have grown in regard can be advantageous.
to its effects on intestinal physiology (Englyst & Macfarlane, 1986; Different factors are involved in the production of unripe banana
Topping & Clifton, 2001; Davis & Milner, 2009; Conlon et al., 2012) as flour (UBF), as banana ripening stage and drying process. The carbohy-
well as its ability to promote a reduced glycemic response (Anderson drate composition of bananas is largely altered during ripening as the
et al., 2010; Klosterbuer, Thomas, & Slavin, 2012) and increased insulin starch reserve is hydrolyzed to soluble sugars (Cordenunsi & Lajolo,
sensitivity (Robertson, Bickerton, Dennis, Vidal, & Frayn, 2005). A cause- 1995). As a large proportion of the starch in unripe bananas is in the
and-effect relationship exists between the intake of RS as a replacement form of resistant starch, it is important to assess the proper stage of
for a portion of the available starch in baked goods and a reduction in fruit ripening to produce flour with a high RS content (Englyst &
post-prandial glycemic response (EFSA, 2011). Increased insulin sensi- Cummings, 1986; Zhang, Whistler, BeMiller, & Hamaker, 2005). The
tivity has also been observed in studies of prolonged RS ingestion (4 process generally employed for unripe banana flour (UBF) drying uti-
lizes a tray dryer or hot-air dryer (Haslinda, Cheng, Chong, & Noor
Aziah, 2009). The amount of RS in UBF has been reported in early stud-
⁎ Corresponding author at: Av. Prof. Lineu Prestes, 580, Bloco 14, CEP 05508-000 São
ies; Englyst, Kingman, and Cummings (1992) evaluated UBF samples
Paulo, SP, Brazil. and found 57.0% RS, whereas Faisant et al. (1995) reported a value of
E-mail address: [email protected] (E.W. Menezes). 54.0%. In Brazil, Tribess et al. (2009) established a process to produce

http://dx.doi.org/10.1016/j.foodres.2015.11.016
0963-9969/© 2015 Elsevier Ltd. All rights reserved.
204 F.A. Hoffmann Sardá et al. / Food Research International 81 (2016) 203–209

UBF with a high RS content (51.0 ± 7.3%), whereby the process temper- with a 12 mm depth (Ditchfield, 2004) and a pre-test speed of 5 mm/s.
ature remains below gelatinization. The analysis of TSS, expressed in °Brix, according to AOAC 932.12 method-
Some authors have suggested the use of unripe banana peel due to ology (Horwitz & Latimer, 2006), was determined using a digital refrac-
its higher content of antioxidants (Haslinda et al., 2009). However, it tometer (DR201-95, Krüss-Optronic GmbH, Hamburg, Germany) and a
is important to consider that pesticide residues might be found in com- portable refractometer (model 950.032 B-ATC, Alla, France).
mercial banana peels (Aurore, Parfait, & Fahrasmane, 2009), which may
result in food safety issues.
2.2.2. Chemical composition
The application of UBF in foods has been evaluated in several studies.
All of the chemical analyses were performed in triplicate, and the re-
A lower glycemic index and load as well as higher fermentability and
sults are expressed in %, equivalent to g/100 g, of dry (d.w.) and fresh
short-chain fatty acid (SCFA) production in vitro have been reported for
weight (f.w.). All samples were ground to particles b0.250 mm prior
cereal bars with UBF compared to a control cereal bar (Santos, 2010). Clin-
to the chemical analyses. The determination of the moisture content
ical trials with UBF have demonstrated hunger/satiety positive responses
of the samples was performed in a vacuum oven at 70 °C (AOAC
when compared with controls, as well as improvement in bowel function
920.151). The protein content was determined by the total nitrogen
in healthy volunteers (Giuntini et al., 2015; Menezes et al., 2010).
present using the micro-Kjeldahl technique (AOAC 960.52) and consid-
UBF has been applied (pilot scale) in several food products, such as ce-
ering a conversion factor 6.25; the lipid content was determined using
real bars (Santos, 2010), crackers (Fasolin, Almeida, Castanho, & Netto-
the Soxhlet method (AOAC 920.39) and the ash content by calcination
Oliveira, 2007), pasta (Agama-Acevedo et al., 2009), and noodles (Choo
in a muffle furnace at 550 °C to a constant weight (AOAC 923.03)
& Aziz, 2010). In addition, commercial UBF can be purchased from natural
(Horwitz & Latimer, 2006). RS was quantified based on the AOAC
product stores and on the Internet in both Brazil (http://verdenutri.com.
2002.02 method (McCleary, McNally, & Rossiter, 2002), and the method
br, http://www.natue.com.br) and abroad (http://www.bananaflour.
of Cordenunsi and Lajolo (1995) was used to quantify the total starch
com/, www.bananaflour.com.au). In a globalized food environment, the
(TS). The amount of free glucose was determined using an enzymatic
current strategies available for monitoring and detecting economically
method (glucose oxidase/peroxidase/ABTS) (Bergmeyer & Bernet,
motivated adulteration have relative strengths and weaknesses, and
1974). The following reference materials were used: RS of corn and po-
new approaches have been recommended (Manning & Soon, 2014). To
tato (Megazyme International Ireland Ltd., Wicklow, Ireland, Megazyme
date, there is no specific legislation for UBF identity and quality standards
K-RSTCL) and cooked carioca beans (in house) were used to determine
available in Brazil, Canada, Australia or the United States, among other
RS, and starch from potato (Sigma S-2004) was used for the determina-
countries. Although commercial UBF has been used in a clinical trial
tion of TS. The concentration of available starch (AS) was obtained from
with metabolic syndrome volunteers (Tavares da Silva et al., 2014), its
the difference between the total and resistant starch (AS = TS − RS).
RS content has not been assessed.
Soluble sugars (SS) were extracted and qualified according to a previous
Therefore, to fill this gap, the purpose of this work was to assess and
report (Cordenunsi, Shiga, & Lajolo, 2008). Glucose, fructose and sucrose
propose characterization parameters for commercially available UBF.
(Sigma, Chemical CO, Saint Louis, MO, USA) were used as reference ma-
terials. The total available carbohydrate content was calculated by the
2. Material and methods
sum of SS and AS. Dietary fiber (DF) was quantified by the enzymatic-
gravimetric method according to AOAC method 991.43 (Horwitz &
2.1. Unripe banana flour (UBF) samples
Latimer, 2006), with the modifications proposed by McCleary and
Rossiter (2004) to exclude RS from DF. Total DF (TDF) was determined
Standard UBF was prepared from the pulp of unripe bananas, Musa
as the sum of DF (without RS) and RS analyzed.
acuminata (group AAA), sub-group Cavendish (called ‘Nanicão’ in
Brazil), not subjected to a ripening chamber. The material was obtained
from Vale do Ribeira and marketed in CEAGESP/SP, maturity stage I, and 2.3. Light microscopy (LM)
UBF production at the semi-industrial scale was performed according to
a proposed process (Tribess et al., 2009). Three different batches were Five reference materials were used for LM: corn starch (Sigma
produced. Standard Peel UBF was produced in the same manner as Stan- S5296); potato starch (Sigma S-2004) and in-house UBF, Standard
dard UBF with the peel retained. Cooked UBF was produced in the same UBF, Standard Peel UBF and Standard cooked UBF described above.
manner as Standard UBF but was heated to 120 °C for 1 min before dry- The samples were prepared as indicated in American and Brazilian
ing process as a standard of gelatinized starch. The UBF starch gelatini- Pharmacopeia (American Pharmacopeia, 2014; Brasil, 2010) and ana-
zation temperature ranges from 67.9 to 71.3 °C (Rayo et al., 2015; lyzed using a light microscope (Olympus model CHS, Olympus Optical
Tribess et al., 2009). The commercial unripe banana flours (n = 12) ex- Co Ltd., Japan). Qualitative analysis of structures was performed with
amined were produced in the states of São Paulo (6), Rio Grande do Sul the preparation of both Lugol and glycerin slides for each sample. Quan-
(2), Santa Catarina (2), Bahia (1) and Minas Gerais (1) and were pur- titative analysis of yellow structures and gelatinized starch was per-
chased in stores that specialize in natural products. The commercial formed using 10 μl of the prepared solution for each sample (50 mg of
UBFs were labeled A to L. sample + 1.5 ml Lugol's solution). Counts were performed in the 4 pe-
ripheral squares of a Neubauer chamber (Mirrored; 0.100 mm depth;
2.2. Physical–chemical analyses 0.0025 mm2; New Optics), and the average was calculated (Umerie &
Ezeuzo, 2000).
2.2.1. Unripe banana maturation stage For the histochemical analysis, different treatments were performed
The unripe banana maturation stage, before UBF production, was to identify the composition of the yellow structures: Calcofluor white
assessed using 3 parameters: total soluble solids content (TSS), peel for cellulose; Sudan IV for lipids; Ferric chloride for phenolic com-
color and pulp firmness. Bananas have 7 ripening stages assessed by pounds; and Vanillin-hydrochloric acid for tannins (Demarco, 2014;
peel color. The color assessment was performed via visual assessment Demarco, de Moraes Castro, & Ascensão, 2013; Ruzin, 1999). Observa-
using randomly chosen bananas according to Von Loesecke's compara- tions and photomicrographs were achieved using a Leica DMLB light
tive standard (PBMH & PIF, 2006). The analysis of firmness was per- microscope equipped with a 100-W mercury vapor lamp, a UV filter
formed using a TA.XT2i Texture Analyzer (EqEP08TP) (Stable Micro cube (excitation filter BP340–380, dichromatic mirror RKP400, suppres-
Systems Ltd., Godalming, Surrey, UK) with a 6-mm diameter probe. For sion filter LP425) and a blue filter cube (excitation filter BP420–490, di-
the analysis, unpeeled bananas were penetrated at three sites (right chromatic mirror RKP510, suppression filter LP515). The histochemical
and left extremities and central). The penetration speed was 1 mm/s analysis flowchart is illustrated in Fig. S1 (Supplementary material).
 F.A. Hoffmann Sardá et al. / Food Research International 81 (2016) 203–209 205

2.4. Statistical analysis E) exhibiting a very high TS content: 92.87 and 92.13%, respectively
(Table 1). The value obtained for Standard UBF was approximately
All results are expressed as the mean ± standard deviation (SD). 80.0% TS, a value similar to the 76.77% (d.w.) previously reported
Kolmogorov–Smirnov's Test was applied to verify whether the data (Menezes et al., 2011). UBF obtained by the lyophilization of eight
were parametric or non-parametric. Parametric data were compared Brazilian banana varieties had a TS content that ranged from 61.00%
using three or more measurements, and one-way analysis of variance (Mysore) to 76.50% (‘Nanica’) (Mota, Lajolo, Cordenunsi, & Ciacco,
(one-way ANOVA) with a post hoc Tukey test was used to determine 2000).
statistically significant differences. Correlation analyses were performed A wide range of available starch contents were found among the
using Pearson's test. Mann–Whitney's test was performed for non- commercial UBFs, ranging from 2.50 to 88.14% (Table 1). Three hypoth-
parametric data, and Spearman's coefficient was used to verify data cor- eses emerged based on this wide range of RS and AS content: (1) the
relations. P values b 0.05 were considered to indicate significant differ- flours were produced from bananas in an advanced ripening stage;
ences. Multivariate statistical cluster analysis was used to assess the (2) the processing was not effective in maintaining RS; or (3) the flours
sub-groupings of the UBFs analyzed. The different UBFs were adopted were not produced from unripe banana pulp. Thus, in addition to light
as observations (rows) and the food components as variables (col- microscopy, analyses of SS and other compounds were performed to in-
umns). In the principal component analysis (PCA), Eigenvalues greater vestigate these hypotheses.
than one were separated to show the projection of the foods in the Among the sugars analyzed individually (glucose, fructose and su-
plan (1 × 2). crose), sucrose was the major contributor to the variation observed in
The statistical analyses were performed using the IBM® SPSS® Sta- the results for SS (Table 2). Regarding the SS value, Standard UBF was
tistics 20.0 software (IBM Corp, Armonk, NY, USA). 0.66% (d.w.), Standard Peel UBF was 2.64% (d.w.), and the commercial
UBF values ranged from 0.05 to 5.66% (d.w.). The two highest values ob-
3. Results and discussion served were for brands F (5.67%) and D (5.66%). These results indicate
either the possible use of fruits in more advanced ripening stages or
3.1. Standard UBF peel utilization. The pulps of unripe fruits typically contain 1–2% sugar,
a percentage that increases to 15–20% when the banana is ripe (Mota
To ensure that the appropriate ripening stage for the cultivar was et al., 2000). SS analysis of the banana peels also showed an increase ac-
used for the production of Standard UBF and UBF Peel, three parameters cording to the ripening stage and variety, which ranged from 0.9–4.3%
were assessed. The bananas were classified in ripening stage I (green) (stage 1) to 23.2–42.2% (stage 7) (d.w.) (Happi Emaga, Andrianaivo,
by color analysis, and the average value for firmness was 31.4 (29.9– Wathelet, Tchango, & Paquot, 2007).
34.7) N. The value for TSS of unripe pulp was 2.8 (1.7–3.2) °Brix.
These values are in agreement with those reported by Tribess et al. 3.3. Unripe banana flour chemical composition
(2009) in bananas of the same species and subgroup in the months of
July/2006 (26.4–27.6 N and 1.2–3.4 °Brix) and May/2007 (29.4–33.5 N Major variations were noted in the chemical analyses performed for
and 1.2 and 2.9 °Brix). According to Ditchfield (2004), consistency the commercial UBF. The largest variation occurred in the ash content,
reaches 14 ± 2 N and TSS 7 ± 1 °Brix in ripening stage 2. Therefore, it with values ranging from 0.28–5.87% (f.w.) (Table 3). Standard UBF
was possible to characterize the unripe banana used to prepare stan- showed an ash content of 2.45% and Standard Peel UBF 4.27%. Haslinda
dard UBFs as being in ripening stage I. et al. (2009) prepared UBF from Awak ABB bananas with and without
Because it contains a high content of RS and a low content of SS, UBF the peel and found ash contents of 2.46% and 3.58%, respectively; the
is considered to be a source of RS. Additionally, because it can provide UBF produced by Fasolin et al. (2007) from Musa cavendish ‘anã’ pulp,
benefits to human health, UBF has the potential to be used as a function- ‘Nanica’ cultivar, had an ash content of 2.62%; and Borges, Pereira, and
al ingredient when consumed on a regular basis (Menezes et al., 2011; Lucena (2009) found 2.59% ash in UBF produced from ‘Prata’ cultivar
Giuntini et al., 2015). The Standard UBF and Peel UBF produced in this pulp. These results are compatible with Standard UBF and some of the
study showed a RS content of 57.48% and 42.25% (d.w.), respectively, commercial UBFs. In addition, flour produced with the banana peel of 6
values that were similar to those found by Englyst et al. (1992) and different varieties presented an ash content that ranged from 6.4 to
Faisant et al. (1995). In a previous trial that used the same banana culti- 12.8% (Happi Emaga et al., 2007). Thus, there is evidence that the pres-
var and production process, UBF was found to be 49.00% RS, 7.20% die- ence of banana peel in UBF tends to increase the ash content. Commercial
tary fiber and 1.81% SS (d.w.) (Menezes et al., 2011). UBF brands B and E showed ash values of 0.28 and 0.35%, respectively,

3.2. Unripe banana flour carbohydrate parameters Table 1

Total starch (TS), resistant starch (RS) and available starch (AS) content (%), dry weight, of
The 12 commercial UBFs used in this study were manufactured by unripe banana flours (UBFs).
different companies located in five Brazilian states. All of the UBFs Unripe banana flours TS (%) RS (%) AS (%)
were purchased in stores specialized in the commercialization of ‘health
Standard UBF 80.02 ± 1.44a 57.48 ± 1.12a 22.82
food products’ and presented health benefit claims on their packages. Standard Peel UBF 74.50 ± 0.89abf 42.25 ± 1.61bc 34.23
However, none of the UBF labels mentioned registration as a functional UBF A 70.00 ± 0.21bdg 43.55 ± 1.23b 26.43
ingredient, and there was no information about the banana cultivar or UBF B 92.87 ± 2.17c 9.99 ± 0.30d 82.88
whether pulp or peel was used for UBF production. UBF C 71.19 ± 3.47b 33.79 ± 0.10em 37.40
UBF D 76.00 ± 1.44a 55.23 ± 0.11a 20.77
Regarding RS content (d.w.), only three commercial UBF brands
UBF E 92.13 ± 0.24c 3.98 ± 0.89f 88.13
showed values above 40.0%, which was compatible with the results UBF F 67.94 ± 0.58d 13.59 ± 0.32gd 54.32
found for Standard UBF (57.4%) and Standard Peel UBF (42.25%) and UBF G 64.69 ± 0.82d 62.19 ± 1.16h 2.50
those in the literature (Faisant et al., 1995; Tribess et al., 2009; UBF H 68.92 ± 5.67bd 27.22 ± 0.43i 41.70
UBF I 64.18 ± 4.0bd 30.66 ± 1.86ei 33.52
Menezes et al., 2011; Pelissari, Andrade-Mahecha, Sobral, & Menegalli,
UBF J 76.80 ± 1.46af 26.84 ± 0.55i 49.96
2012). Nine of the UBF brands had a composition of less than 40.0% UBF K 78.24 ± 3.8af 37.44 ± 0.84mn 40.80
RS, and two brands (B and E) showed values less than 10.0%, which sug- UBF L 75.80 ± 1.17afg 38.26 ± 0.93nc 37.54
gests a possible inappropriate composition for those products (Table 1). Results expressed as mean ± standard deviation.
The results of total starch quantification (d.w.) revealed that all the Different letters represent significant difference among data from the same column, differ-
samples had a content higher than 60.0%, with two brands (B and ent products (p b 0.05).
206 F.A. Hoffmann Sardá et al. / Food Research International 81 (2016) 203–209

Table 2 the present study, Standard Peel UBF showed a lipid content of 1.20%,
Glucose, fructose, sucrose and soluble sugars content (%), dry weight, of unripe banana whereas UBFs produced from general banana pulp had a maximum
flours (UBFs).
lipid content of 1% (Haslinda et al., 2009) (Borges et al., 2009)
Unripe banana Glucose (%) Fructose (%) Sucrose (%) Soluble sugars⁎ (Pelissari et al., 2012), with UBFs produced with peel having the highest
flours (%) lipid content. Additionally, Haslinda et al. (2009) reported lipid contents
Standard UBF 0.16 ± 0.05a 0.17 ± 0.01ag 0.35 ± 0.02ac 0.66 ± 0.03a of 0.94 and 1.94% for UBF without peel and with peel, respectively.
Standard Peel UBF 0.30 ± 0.01ab 0.39 ± 0.02bfg 1.96 ± 0.02b 2.64 ± 0.03b The total dietary fiber (TDF) content of Standard UBF was 62.50% and
UBF A 0.20 ± 0.10a 0.18 ± 0.04a 0.36 ± 0.03ac 0.74 ± 0.13a
51.52% for Standard Peel UBF. Five of the brands reached a TDF content of
UBF B 0.28 ± 0.03ab 0.46 ± 0.02b 0.27 ± 0.06a 1.01 ± 0.04ag
UBF C 0.42 ± 0.01b 0.40 ± 0.01b 0.94 ± 0.01d 1.76 ± 0.03c 50%, whereas the other 7 brands remained below this level, with Brands B
UBF D 1.31 ± 0.18c 1.89 ± 0.08c 2.46 ± 0.09e 5.66 ± 0.38d and E showing a very low content: 10.46% and 4.44%, respectively.
UBF E 0.04 ± 0.01a 0.02 ± 0.01a 0.00 ± 0.00f 0.05 ± 0.01e In this study, variations were observed regarding the chemical com-
UBF F 0.80 ± 0.01d 1.23 ± 0.01d 3.64 ± 0.03g 5.67 ± 0.03d position of the UBFs sold on the Brazilian market in relation to their AS,
UBF G 1.01 ± 0.08e 1.61 ± 0.04e 2.26 ± 0.12h 4.88 ± 0.12f
UBF H 0.06 ± 0.01a 0.32 ± 0.02ab 0.71 ± 0.03i 1.09 ± 0.05gh
SS and ash contents. Those variations were especially notable in relation
UBF I 0.08 ± 0.01a 0.15 ± 0.04a 0.53 ± 0.03j 0.76 ± 0.08ag to RS, the functional component of UBF.
UBF J 0.10 ± 0.01a 0.12 ± 0.01a 0.43 ± 0.04cj 0.65 ± 0.05a
UBF K 0.20 ± 0.01a 0.32 ± 0.03bfg 1.11 ± 0.07k 1.63 ± 0.11c
UBF L 0.14 ± 0.01a 0.24 ± 0.01af 1.05 ± 0.02dk 1.43 ± 0.02ch
3.4. Light microscopy (LM)
Results expressed as mean ± standard deviation.
Different letters represent significant difference among data from the same column, differ- Due to the high variability observed in the content of total and resis-
ent products (p b 0.05).
⁎ Soluble sugars calculation: Glucose + fructose + sucrose (same triplicate). tant starch in commercial UBFs, an analysis by light microscopy was
performed to evaluate the characteristics of those starches and the pres-
ence of other compounds.
values that are extremely low and different from Standard UBF and the Qualitative comparison of the 12 commercial UBFs with the standards
other brand (Table 3). (Standard UBF, corn starch, potato starch) revealed that two brands, B
Moisture in Standard UBF was an average of 6.34% but was 7.16 in (Fig. 1.7) and E (Fig. 1.8), contained no typical banana starch, even
Standard Peel UBF and ranged from 3.7 to 10.6% in the commercial when they were compared with 5 different types of banana starch
UBFs (Table 3); these values are within the maximum limit of 15% (Peroni-Okita et al., 2010; Soares et al., 2011). In fact, these products pre-
established by the Brazilian legislation under resolution RDC 263 sented characteristics similar to corn starch (Fig. 1.3), with polyhedral
(BRASIL, 2005). Moisture values identified in UBFs produced with ba- granules, a round, cleft or stellar hilum, and size of 14–20 μm (American
nanas of the same cultivar and following the same drying process Pharmacopeia, 2014; Brasil, 2010), as shown in Fig. 1. Those two brands
were 9.62% (Dan, 2011) and 4.00% in July 2006 and 7.60% in May 2007 also showed a reduced RS content, 10% and 4%, respectively; a high con-
(Tribess et al., 2009). A UBF produced under similar conditions and tent of TS (Table 1); reduced protein and ash contents (Table 3); and a
using the ‘Prata’ cultivar showed a moisture value of 3.30% (Borges white color. Therefore, it was possible to establish that these products
et al., 2009). are not actually UBF but, rather, another cereal starch. Noronha et al.
The protein analysis revealed 3.72% for Standard UBF and 5.51% for (2008) have also applied LM to identify starch and to determine success-
Standard Peel UBF and ranged between 4.01 and 5.79% for the commer- fully its distribution in starch-containing imitation cheese.
cial UBFs (f.w.), except for 2 brands (B and E), which showed much The other 10 commercial UBFs, Standard UBF and Peel UBF showed
lower results: 0.64 and 0.00%, respectively. The protein content of UBF starch characteristics of banana, with the presence of spherical and
produced using other banana cultivars was 4.50% for ‘Prata’ (Borges elongated forms with no regularity, growth rings and a size of 10–
et al., 2009) and 6.28% and 6.77% for Awak ABB with and without the 70 μm (Peroni-Okita et al., 2013; Peroni-Okita et al., 2010); moreover,
peel, respectively (Haslinda et al., 2009). Thus, protein content values no contamination or presence of any other starch type was found. The
between 3 and 6% appear to be characteristic of UBF. qualitative assessment of these characteristic UBFs (with a banana
Regarding the lipid content (f.w.), brand E had the lowest value starch identity) revealed the presence of atypical structures of a yellow
(0.04%) and brand G the highest value (2.07%). Standard UBF showed color and the presence of gelatinized starch. Based on that observation,
a mean of 0.38% for lipids. Dan (2011) identified 0.10% lipids in a UBF a quantitative analysis of those 10 commercial UBFs was performed
sample, and the value found by Menezes et al. (2011) was 0.89%. In using a Neubauer chamber.

Table 3
Chemical composition (%), fresh weight, and energy value (kJ/kcal) in unripe banana flours (UBFs).

Unripe banana flours Moisture (%) Protein (%) Lipids (%) Ash (%) Total dietary fiber⁎ Available carbohydrates⁎⁎
a a a
Standard UBF 6.34 ± 0.74 3.72 ± 0.12 0.38 ± 0.04 2.45 ± 0.07 62.50 21.73
Standard Peel UBF 7.16 ± 0.32 5.51 ± 0.19bc 1.20 ± 0.08b 4.27 ± 0.07b 51.52 32.46
UBF A 5.08 ± 0.20 5.24 ± 0.11b 1.36 ± 0.16bc 4.46 ± 0.01c 54.55 25.86
UBF B 9.12 ± 0.06 0.64 ± 0.07d 0.14 ± 0.05d 0.28 ± 0.01d 9.33 76.24
UBF C 8.75 ± 0.06 5.23 ± 0.02bg 1.54 ± 0.16cfg 4.90 ± 0.02e 46.24 35.76
UBF D 10.60 ± 0.17 4.01 ± 0.06a 0.92 ± 0.02h 3.81 ± 0.01f 59.90 23.91
UBF E 7.67 ± 0.22 0.00 ± 0.00e 0.06 ± 0.03d 0.35 ± 0.01d 4.52 81.43
UBF F 4.25 ± 0.26 4.32 ± 0.01f 1.50 ± 0.02ck 4.51 ± 0.06cg 27.44 57.44
UBF G 5.26 ± 0.15 4.65 ± 0.16f 2.08 ± 0.01m 5.17 ± 0.06h 76.35 6.38
UBF H 3.65 ± 0.08 4.53 ± 0.13f 1.30 ± 0.07bk 4.89 ± 0.03e 39.67 41.16
UBF I 6.65 ± 0.32 5.56 ± 0.30bc 1.60 ± 0.02f 5.87 ± 0.05i 43.53 31.99
UBF J 4.14 ± 0.24 5.79 ± 0.24c 1.35 ± 0.04bgk 4.97 ± 0.05ek 39.26 48.85
UBF K 5.54 ± 0.10 5.04 ± 0.07g 1.21 ± 0.02b 5.06 ± 0.03hk 47.76 40.08
UBF L 7.66 ± 0.11 5.41 ± 0.12bg 1.33 ± 0.08bk 4.64 ± 0.03g 46.62 35.96

Results expressed as mean ± standard deviation.

Different letters represent significant difference among data from the same column, different products (p b 0.05).
⁎ Total dietary fiber calculation: (Resistant starch + dietary fiber).
⁎⁎ Available carbohydrates calculation: (Available starch + soluble sugars).
 F.A. Hoffmann Sardá et al. / Food Research International 81 (2016) 203–209 207

Fig. 1. Image obtained by light microscopy to illustrate the investigated structures in Standard UBF, Standard Peel UBF, Standardcooked UBF, corn starch reference material and selected
commercial UBFs: B, D, E, F and J. Slides prepared with Lugol's solution.

In relation to the above-mentioned yellow structures, Standard UBF unripe banana pulp as well as part of the peel were used to the flour
is different from Standard Peel UBF with the latter having a greater in produce brands A, C, F, G, H, I, J, K and L, though there is no clear indi-
amount of these structures. These results suggest that these structures cation of this addition on the labels.
are likely to be more abundant in the peel (Table 4). This result was ver- Regarding the presence of gelatinized starch, Standard UBF showed
ified by a positive correlation between the presence of yellow structures no significant difference in comparison with Peel UBF. However, the
and the contents of lipids (r = 0.870, p b 0.001), ash (r = 0.812, p = cooked UBF (Fig. 1.4) that was subjected to heating at 120 °C showed
0.001) and proteins (r = 0.704, p = 0.005), which are higher in abun- 100% gelatinized starch. When comparing the 10 UBF brands, a signifi-
dance in the peel, and a negative correlation with the content of total cant difference in relation to Standard UBF was found only for brands
starch (r = −0.761, p = 0.002), which is more abundant in the pulp. F (Fig. 1.6) and J (Fig. 1.9) (Table 4), thus indicating that the drying pro-
Histochemical analysis performed to examine the yellow structures cess possibly occurred at higher temperatures, which can also be evi-
suggested that they are composed of phenolic compounds and tannins denced by the low content of RS of only 13.6% and 26.8%, respectively
(Fig. S1— Supplementary Material). Tanniferous cells, laticifers and co- (Table 1). A negative correlation between RS and gelatinized starch
agulated latex with tannins are common structures observed in banana was also observed (r = − 0.676, p = 0.008). It is worth mentioning
peel (Tee, Ding, & Rahman, 2011). The total phenolic content of banana that none of the brands identified the unripe banana cultivar used.
flours with and without peel were reported to be 1402 mg and 747 mg
catechol equivalents per 100 g sample, respectively (Haslinda et al., 3.5. Clustering analysis
2009). When comparing the 10 UBF brands in the present study, nine
showed a significantly higher quantity of yellow structures than Stan- The principal component analysis (PCA) applied to the 12 commer-
dard UBF and a similar number to Standard Peel UBF (brands A, C, G, cial UBFs, Standard UBF and Peel UBF enabled the identification of and
H, I, J, K and L); UBF brand F (Fig. 1.6) had more yellow structures separation into three subgroups, as well as a view of the groups in a p-
than Standard Peel UBF (Fig. 1.2), though D (Fig. 1.5) showed no signif- dimensional space (Fig. 2). The first two components (Principal compo-
icant difference from Standard UBF (Fig. 1.1). It can be assumed that nent 1 (PC1) and 2 (PC2); Eigenvalues N 1) were selected for analysis,
and 68.83% (PC1) and 16.19% (PC2), respectively, were obtained with
a total variation of 85.2%. PC1 is represented by the dietary fiber
Table 4 (0.979), ash (0.949) and lipid (0.939) contents, whereas PC2 is repre-
Mean value of characteristic yellow structures and gelatinized starch (units/100 μg) pres-
sented by the resistant starch (− 0.746) content. Three clusters were
ent in unripe banana flours (UBFs).
obtained:
Unripe banana Yellow structures Gelatinized starch Cluster 1 (UBF brands B and E) — containing extremely low RS, DF,
flours (units/100 μg) (units/100 μg)
ash and lipid contents and corresponding to the two samples composed
Standard UBF 23 ± 18a 6 ± 10a of corn starch with no presence of unripe banana starch (Fig. 1).
Standard Peel UBF 376 ± 59b 2 ± 6a Cluster 2 (Standard UBF, UBF brands D and G) — containing a high RS
UBF A 465 ± 60b 10 ± 13a
content and higher DF, ash and lipid contents and corresponding to the
UBF C 452 ± 86b 36 ± 22a
UBF D 121 ± 41a 18 ± 13a standard sample and two commercial UBFs that were similar.
UBF F 574 ± 103c 48 ± 22b Cluster 3 (UBF brands A, C, F, H, I, J, K and L, Standard Peel UBF) —cor-
UBF G 419 ± 84b 3 ± 8a responding to the standard peel sample and commercial UBF samples
UBF H 363 ± 87b 28 ± 24a
with a low RS content and high DF, ash and lipid contents.
UBF I 342 ± 108b 10 ± 16a
UBF J 308 ± 63b 243 ± 82c Based on chemical and microscopic analyses of the 12 commercial
UBF K 419 ± 72b 8 ± 10a UBFs, the following conclusions can be drawn: 1) two UBF brands (B
UBF L 346 ± 93b 2 ± 6a and E) were not produced using unripe banana but rather with cereal
Results expressed as mean ± standard deviation (triplicate). Different letters represent starch; 2) two brands (F and J) had a low content of RS, which is likely
significant difference among data from the same column: different products (p b 0.05). due to excessive heating during manufacturing, as evidenced by the
208 F.A. Hoffmann Sardá et al. / Food Research International 81 (2016) 203–209

Vegetal — Instituto de Biociências — Universidade de São Paulo) for as-

sistance with the histochemistry analysis.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.foodres.2015.11.016.

References
Agama-Acevedo, E., Islas-Hernandez, J. J., Osorio-Díaz, P., Rendón-Villalobos, R., Utrilla-
Coello, R. G., Angulo, O., & Bello-Pérez, L. A. (2009). Pasta with unripe banana flour:
Physical, texture, and preference study. Journal of Food Science, 74(6), S263–S267.
http://dx.doi.org/10.1111/j.1750-3841.2009.01215.x.
Almeida, C. O. (2012). Economia. In Embrapa (Ed.), Coleção 500 perguntas 500 respostas: ba-
nana. O produtor pergunta, a embrapa responde (pp. 2011–2014). Brasília: Embrapa.
American Pharmacopeia. (2014). Baltimore, MD, USA: United Book Press, Inc.
Anderson, G. H., Cho, C. E., Akhavan, T., Mollard, R. C., Luhovyy, B. L., & Finocchiaro, E. T.
(2010). Relation between estimates of cornstarch digestibility by the Englyst
in vitro method and glycemic response, subjective appetite, and short-term food in-
take in young men. American Journal of Clinical Nutrition, 91(4), 932–939. http://dx.
doi.org/10.3945/ajcn.2009.28443.
Aurore, G., Parfait, B., & Fahrasmane, L. (2009). Bananas, raw materials for making proc-
essed food products. Trends in Food Science & Technology, 20(2), 78–91. http://dx.
doi.org/10.1016/j.tifs.2008.10.003.
Bergmeyer, H., & Bernet, E. (1974). Determination of glucose with oxidase and peroxidase
Fig. 2. Principal components analysis (PCA) of Standard UBF, Standard Peel UBF and 12 Enzymatic Analysis (Second edition ). New York: Academic Press, 1205–1215.
commercial UBFs from data of chemical composition (proteins (%), lipids (%), ash (%), re- Birt, D. F., Boylston, T., Hendrich, S., Jane, J. L., Hollis, J., Li, L., ... Whitley, E. M. (2013). Re-
sistant starch (%), available starch (%), dietary fiber (%), soluble sugars (%)) and microsco- sistant starch: Promise for improving human health. Advances in Nutrition, 4(6),
py (yellow structures (units/100 μg)). 587–601. http://dx.doi.org/10.3945/an.113.004325.
Borges, A., Pereira, J., & Lucena, E. (2009). Caracterização da farinha de banana verde.
Ciência e Tecnologia de Alimentos, 29(2), 333–339. http://dx.doi.org/10.1590/S0101-
greater presence of gelatinized starch and the significant difference 20612009000200015.
Brasil (2005). Agência Nacional de Vigilância Sanitária. RDC n° 263, de 22 de setembro de
when compared with Standard UBF; and 3) five brands (C, H, I, K and
2005. Regulamento técnico para produtores de cereais, amidos, farinhas e farelos. (263,
L) had a RS content that ranged between 27.2 and 38.3%, and the most de 22 de setembro de 2005). Brasília: Diário Oficial da União.
probable causes for the reduced values for these five brands were the Brasil (2010). Farmacopeia Brasileira Vol. 2. Retrieved from http://www.anvisa.gov.br/
hotsite/cd_farmacopeia/index.htm
proportion of the peel to the pulp, as evidenced by a higher ash content
Choo, C. L., & Aziz, N. A. A. (2010). Effects of banana flour and β-glucan on the nutritional
and lower TS content, as well as by the presence of gelatinized starch at and sensory evaluation of noodles. Food Chemistry, 119(1), 34–40. http://dx.doi.org/
a higher, albeit not significant, proportion. 10.1016/j.foodchem.2009.05.004.
Only three brands (A, D and G) exhibited a RS content higher than Conlon, M. A., Kerr, C. A., McSweeney, C. S., Dunne, R. A., Shaw, J. M., Kang, S., ... Topping, D.
L. (2012). Resistant starches protect against colonic DNA damage and alter microbio-
40% (similar to Standard UBF) and the potential to be used as a function- ta and gene expression in rats fed a Western diet. Journal of Nutrition, 142(5),
al ingredient due to the preserved content of RS. However, it was not 832–840. http://dx.doi.org/10.3945/jn.111.147660.
possible to determine whether those results were due to a controlled Cordenunsi, B. R., & Lajolo, F. M. (1995). Starch breakdown during banana ripening — Su-
crose synthase and sucrose-phosphate synthase. Journal of Agricultural and Food
manufacturing process or whether they occurred randomly. It is note- Chemistry, 43, 347–351. http://dx.doi.org/10.1021/jf00050a016.
worthy that UBFs A and G had higher ash and lipid contents in addition Cordenunsi, B. R., Shiga, T. M., & Lajolo, F. (2008). Non-starch polysaccharide composition of
to the presence of yellow structures, thus suggesting the possible use of two cultivars of banana (Musa acuminata L.: cvs Mysore and Nanicão). Carbohydrate
Polymers, 71(1), 26–31. http://dx.doi.org/10.1016/j.carbpol.2007.05.009.
unripe banana peel. Dan, M. C. T. (2011). Avaliação da potencialidade da farinha de banana Verde Como
ingrediente funcional: Estudo in vivo e in vitro. (Doctorate thesis) São Paulo (Brasil):
Universidade de São Paulo.
4. Conclusions Davis, C. D., & Milner, J. A. (2009). Gastrointestinal microflora, food components and colon
cancer prevention. Journal of Nutritional Biochemistry, 20(10), 743–752. http://dx.doi.
org/10.1016/j.jnutbio.2009.06.001 C2 - 19716282.
The 12 different unripe banana flours (UBFs) purchased from mar-
Demarco, D. (2014). Secretory tissues and the morphogenesis and histochemistry of
kets and evaluated in terms of their chemical and physical characteris- pollinarium in flowers of asclepiadeae (Apocynaceae). International Journal of Plant
tics showed a lack of standardization and high variability in the Sciences, 175(9), 1042–1053. http://dx.doi.org/10.1086/677947.
Demarco, D., de Moraes Castro, M., & Ascensão, L. (2013). Two laticifer systems in Sapium
content of resistant starch (RS) when compared with Standard UBF
haematospermum — New records for Euphorbiaceae. Botany, 91(8), 545–554. http://
and other standards in the literature. Indeed, only three products dx.doi.org/10.1139/cjb-2012-0277.
showed a RS content within the expected range. The characterization Ditchfield, C. (2004). Estudo do processamento contínuo do purê de banana (Musa
of UBF by adding minimal labeling information (the unripe banana cul- cavendishii Lamb.). (Doctorate thesis) São Paulo (Brasil): Universidade de São Paulo.
EFSA (2011). Scientific Opinion on the substantiation of health claims related to resistant
tivar used, use of pulp and/or peel) and carbohydrate parameters (RS starch and reduction of post-prandial glycaemic responses (ID 681), “digestive health
and soluble sugar content) might help consumers in their choice of a benefits” (ID 682) and “favours a normal colon metabolism” (ID 783) pursuant to Ar-
product, and the use of light microscopy (to identify the starch type) ticle 13(1) of Regulation (EC) No 1924/2006. European Food Safety Authority Journal,
9(4), 2024. http://dx.doi.org/10.2903/j.efsa.2011.2024 (2017 pp.).
could facilitate the performance of food inspection agencies. Englyst, H. N., & Cummings, J. H. (1986). Digestion of the carbohydrates of banana (Musa
paradisiaca sapientum) in the human small intestine. American Journal of Clinical Nu-
trition, 44(1), 42–50.
Acknowledgments Englyst, H. N., & Macfarlane, G. T. (1986). Breakdown of resistant and readily digestible
starch by human gut bacteria. Journal of the Science of Food and Agriculture, 37(7),
699–706. http://dx.doi.org/10.1002/jsfa.2740370717/abstract.
The authors acknowledge financial support from the São Paulo Re- Englyst, H. N., Kingman, S. M., & Cummings, J. H. (1992). Classification and measurement
search Foundation (FAPESP) under grants 2012/24607-9 and 2013/ of nutritionally important starch fractions. European Journal of Clinical Nutrition,
07914-8 and also the scholarships provided by the Coordination for 46(Suppl. 2), S33–S50.
Faisant, N., Buléon, A., Colonna, P., Molis, C., Lartigue, S., Galmiche, J. P., & Champ, M.
the Improvement of Higher Education Personnel (CAPES) and by the
(1995). Digestion of raw banana starch in the small intestine of healthy humans:
National Council for Scientific and Technological Development (CNPq). Structural features of resistant starch. British Journal of Nutrition, 73(1), 111–123.
We thank Fernanda H.G. Peroni-Okita, PhD, and Professor Beatriz R. FAO (2014). Food and agriculture oganiztion of the United Nations. Food and agricultural
Cordenunsi (Department of Food Science and Experimental Nutrition, commodities production.
Fasolin, L. H., Almeida, G. C. D., Castanho, P. S., & Netto-Oliveira, E. R. (2007). Biscoitos
FCF-USP) for assistance with the banana starch microscopy identifica- produzidos com farinha de banana: Avaliações química, física e sensorial. Food
tion. We also thank Professor Diego Demarco (Laboratório de Anatomia Science and Technology (Campinas), 27, 524–529.
 F.A. Hoffmann Sardá et al. / Food Research International 81 (2016) 203–209 209

Fuentes-Zaragoza, E., Riquelm-Navarrete, M., Sánchez-Zapata, E., & Pérez-Álvarez, J. (2010). Pelissari, F. M., Andrade-Mahecha, M. M., Sobral, P. J. d. A., & Menegalli, F. C. (2012). Isolation
Resistant starch as functional ingredient: a review. Food Research International, 43, and characterization of the flour and starch of plantain bananas (Musa paradisiaca).
931–942. Starch/Stärke, 64(5), 382–391. http://dx.doi.org/10.1002/star.201100133/abstract.
Giuntini, E. B., Sardá, F. A. H., Lui, M. C. Y., Tadini, C. C., Lajolo, F. M., & Menezes, E. W. Peroni-Okita, F. H. G., Cardoso, M. B., Agopian, R. G. D., Louro, R. P., Nascimento, J. R. O.,
(2015). Gastrointestinal hormones modulation after a double-blind interventional Purgatto, E., ... Cordenunsi, B. R. (2013). The cold storage of green bananas affects
study with unavailable carbohydrates. Food Research International. http://dx.doi.org/ the starch degradation during ripening at higher temperature. Carbohydrate
10.1016/j.foodres.2015.04.003. Polymers, 96(1), 137–147. http://dx.doi.org/10.1016/j.carbpol.2013.03.050.
Happi Emaga, T., Andrianaivo, R. H., Wathelet, B., Tchango, J. T., & Paquot, M. (2007). Ef- Peroni-Okita, F. H. G., Simão, R. A., Cardoso, M. B., Soares, C. A., Lajolo, F. M., & Cordenunsi,
fects of the stage of maturation and varieties on the chemical composition of banana B. R. (2010). In vivo degradation of banana starch: Structural characterization of the
and plantain peels. Food Chemistry, 103(2), 590–600. http://dx.doi.org/10.1016/j. degradation process. Carbohydrate Polymers, 81(2), 291–299. http://dx.doi.org/10.
foodchem.2006.09.006. 1016/j.carbpol.2010.02.022.
Haslinda, W. H., Cheng, L. H., Chong, L. C., & Noor Aziah, A. A. (2009). Chemical composi- Rayo, L. M., Chaguri e Carvalho, L., Sardá, F. A. H., Dacanal, G. C., Menezes, E. W., & Tadini, C.
tion and physicochemical properties of green banana (Musa acuminata × balbisiana C. (2015). Production of instant green banana flour (Musa cavendischii, var. Nanicão)
Colla cv. Awak) flour. International Journal of Food Science and Nutrition, 60(4), by a pulsed-fluidized bed agglomeration. LWT - Food Science and Technology, 63(1),
232–239. http://dx.doi.org/10.1080/09637480902915525. 461–469. http://dx.doi.org/10.1016/j.lwt.2015.03.059.
Horwitz, W., & Latimer, G. W., Jr. (2006). Official methods of analysis of AOAC international Robertson, M. D., Bickerton, A. S., Dennis, A. L., Vidal, H., & Frayn, K. N. (2005). Insulin-
(18th ed.). Maryland: AOAC International. sensitizing effects of dietary resistant starch and effects on skeletal muscle and adi-
Johnston, K. L., Thomas, E. L., Bell, J. D., Frost, G. S., & Robertson, M. D. (2010). Resistant pose tissue metabolism. American Journal of Clinical Nutrition, 82(3), 559–567.
starch improves insulin sensitivity in metabolic syndrome. Diabetic Medicine, 27(4), Robertson, M. D., Wright, J. W., Loizon, E., Debard, C., Vidal, H., Shojaee-Moradie, F., ...
391–397. http://dx.doi.org/10.1111/j.1464–5491.2010.02923.x. Umpleby, A. M. (2012). Insulin-sensitizing effects on muscle and adipose tissue
Klosterbuer, A. S., Thomas, W., & Slavin, J. L. (2012). Resistant starch and pullulan reduce after dietary fiber intake in men and women with metabolic syndrome. The Journal
postprandial glucose, insulin, and GLP-1, but have no effect on satiety in healthy of Clinical Endocrinology & Metabolism, 97(9), 3326–3332. http://dx.doi.org/10.1210/
humans. Journal of Agricultural and Food Chemistry, 60(48), 11928–11934. http://dx. jc.2012-1513.
doi.org/10.1021/jf303083r. Ruzin, S. E. (1999). Plant microtechnique and microscopy. New York, USA: Oxford Univer-
Maki, K. C., Pelkman, C. L., Finocchiaro, E. T., Kelley, K. M., Lawless, A. L., Schild, A. L., & sity Press.
Rains, T. M. (2012). Resistant starch from high-amylose maize increases insulin sen- Santos, J. F. (2010). Avaliação das propriedades nutricionais de barras de cereais elaboradas
sitivity in overweight and obese men. Journal of Nutrition, 142(4), 717–723. http://dx. com farinha de banana Verde. (Master's thesis) São Paulo: Universidade de São Paulo.
doi.org/10.3945/jn.111.152975. Soares, C. A., Peroni-Okita, F. H., Cardoso, M. B., Shitakubo, R., Lajolo, F. M., & Cordenunsi, B.
Manning, L., & Soon, J. M. (2014). Developing systems to control food adulteration. Food R. (2011). Plantain and banana starches: granule structural characteristics explain the
Policy, 49(Part 1(0)), 23–32. http://dx.doi.org/10.1016/j.foodpol.2014.06.005. differences in their starch degradation patterns. Journal of Agricultural and Food
McCleary, B. V., & Rossiter, P. (2004). Measurement of novel dietary fibers. Journal of AOAC Chemistry, 59(12), 6672–6681. http://dx.doi.org/10.1021/jf201590h.
International, 87(3), 707–717. Tavares da Silva, S., Araújo Dos Santos, C., Marvila Girondoli, Y., Mello de Azeredo, L.,
McCleary, B. V., McNally, M., & Rossiter, P. (2002). Measurement of resistant starch by en- Fernando de Sousa Moraes, L., Keila Viana Gomes Schitini, J., et al. (2014). Women
zymatic digestion in starch and selected plant materials: collaborative study. Journal with metabolic syndrome improve anthropometric and biochemical parameters
of AOAC International, 85(5), 1103–1111. with green banana flour consumption. Nutrición Hospitalaria, 29(5), 1070–1080.
Menezes, E. W., Dan, M. C., Cardenette, G. H., Goñi, I., Bello-Pérez, L. A., & Lajolo, F. M. http://dx.doi.org/10.3305/nh.2014.29.5.7331.
(2010). In vitro colonic fermentation and glycemic response of different kinds of un- Tee, Y. K., Ding, P., & Rahman, N. A. A. (2011). Physical and cellular structure changes of
ripe banana flour. Plant Foods for Human Nutrition, 65(4), 379–385. http://dx.doi.org/ Rastali banana (Musa AAB) during growth and development. Scientia Horticulturae,
10.1007/s11130-010-0190-4. 129(3), 382–389. http://dx.doi.org/10.1016/j.scienta.2011.03.050.
Menezes, E. W., Tadini, C. C., Tribess, T. B., Zuleta, A., Binaghi, J., Pak, N., ... Lajolo, F. M. Topping, D. L., & Clifton, P. M. (2001). Short-chain fatty acids and human colonic function:
(2011). Chemical composition and nutritional value of unripe banana flour (Musa roles of resistant starch and nonstarch polysaccharides. Physiological Reviews, 81(3),
acuminata, var. Nanicão). Plant Foods for Human Nutrition, 66(3), 231–237. http:// 1031–1064.
dx.doi.org/10.1007/s11130-011-0238-0. Tribess, T. B., Hernández-Uribe, J. P., Méndez-Montealvo, M. G. C., Menezes, E. W., Bello-
Mota, R. V. d., Lajolo, F. M., Cordenunsi, B. R., & Ciacco, C. (2000). Composition and functional Perez, L. A., & Tadini, C. C. (2009). Thermal properties and resistant starch content
properties of banana flour from different varieties. Starch/Stärke, 52(2–3), 63–68. http:// of green banana flour (Musa cavendishii) produced at different drying conditions.
dx.doi.org/10.1002/(SICI)1521-379X(200004)52:2/3b63::AID-STAR63N3.0.CO. LWT - Food Science and Technology, 42(5), 1022–1025. http://dx.doi.org/10.1016/j.
Noronha, N., Duggan, E., Ziegler, G. R., Stapleton, J. J., O'Riordan, E. D., & O'Sullivan, M. lwt.2008.12.017.
(2008). Comparison of microscopy techniques for the examination of the microstruc- Umerie, S. C., & Ezeuzo, H. O. (2000). Physicochemical characterization and utilization of
ture of starch-containing imitation cheeses. Food Research International, 41(5), Cyperus rotundus starch. Bioresource Technology, 72(2), 193–196. http://dx.doi.org/
472–479. http://dx.doi.org/10.1016/j.foodres.2008.02.008. 10.1016/S0960-8524(99)00103–0.
PBMH & PIF - Programa Brasileiro Para a Modernização da Horticultura & Produção Zhang, P., Whistler, R. L., BeMiller, J. N., & Hamaker, B. R. (2005). Banana starch: produc-
Integrada de Frutas (2006). Normas de Classificação de Banana. São Paulo: CEAGESP tion, physicochemical properties, and digestibility—a review. Carbohydrate Polymers,
Documentos, 29. Retrieved from http://www.ceagesp.gov.br/wp-content/uploads/ 59(4), 443–458. http://dx.doi.org/10.1016/j.carbpol.2004.10.014.
2015/07/banana.pdf.