Effect of fibre fortification of low FODMAP pasta

Abstract

Irritable bowel syndrome (IBS) is a condition affecting the digestive system and can be triggered by several different factors, including diet. To ease symptoms of IBS, a diet low in fermentable oligo-, di-, monosaccharides and polyols (FODMAPs) is often recommended. Pasta, as a staple food in the Western World, is naturally high in FODMAPs. This study investigates the impact of insoluble and soluble dietary fibre ingredients in low-FODMAPs pasta. The assessment included physicochemical, sensory, and nutritional quality. Soluble fibre strengthened gluten network, which caused a lower cooking loss and a lower release of sugars during in vitro starch digestion. Insoluble fibre interfered with the gluten network development to a higher extent causing a higher sugar release during digestion. This study reveals the most suitable fibre ingredients for the development of pasta with elevated nutritional value and sensory characteristics compared to commercial products on the market. This type of pasta has a high potential of being suitable for IBS patients.

Keywords:

• Dietary fibre
• FODMAPs
• starch digestions
• pasta quality
• food structure

Introduction

Fermentable oligo-, di-, monosaccharides and polyols (FODMAPs) are a group of carbohydrates which are associated with the onset of gastrointestinal symptoms in individuals suffering from irritable bowel syndrome (IBS). FODMAPs are small osmotically active and rapidly fermentable carbohydrates, which cause an increase of luminal gas and reabsorption of water. Therefore, symptoms, such as bloating, abdominal pain or diarrhoea, are triggered (Staudacher et al. 2014). These compounds and their importance for IBS patients have been the focus of research for the last 15 years. FODMAPs occur ubiquitously among food products (Gibson 2017; Varney et al. 2017; Gibson et al. 2020), including cereal-based products, such as bread, pasta and cakes. The indigestible oligosaccharides fructans and α-galactooligosaccharides (GOS) are the main FODMAP carbohydrates typically found in cereals and pulses (Muir et al. 2009; Biesiekierski et al. 2011). A diet with a reduced intake of FODMAPs (known as a low FODMAP diet) has been reported to be efficient in improving IBS symptoms (Gibson 2017; Bellini et al. 2020). Therefore, functional foods which have a low FODMAP content and support the gut health of IBS patients are needed.

Dietary fibres (DF) and FODMAPs share specific characteristics, such as indigestibility and fermentability by gut bacteria (Eswaran et al. 2013; Atzler et al. 2021a). The term DF describes indigestible carbohydrates which are linked with improving gut and general health (Mann and Cummings 2009). Because of the similar characteristics between DF and FODMAPs the low FODMAP diet has been criticised for the elimination of healthy DF. The substitution of eliminated DF with beneficial and IBS suitable DF is crucial. However, maintaining the recommended fibre intake of 25–30 g per day can be challenging (Kendall et al. 2010). The enrichment of foods with DF, including pasta, has been extensively investigated regarding in vitro starch digestibility (Brennan and Tudorica 2008; Gelencsér et al. 2008; Li et al. 2014; Foschia et al. 2015). However, data on the incorporation of “IBS-safe” fibres in food are scarce. Cereal-based fibres, such as wheat bran, were the focus of previous studies investigating the fortification of pasta with DFs (Tudorica et al. 2002 Li et al. 2014; Foschia et al. 2015). However, cereal-based DFs are high in fructans, which are classified as FODMAPs, and therefore, are not suitable for a low FODMAP diet (Ispiryan et al. 2020). Recent research identified DFs that are not classified as FODMAPs and allow IBS patients to ingest fibre without suffering from typical symptoms (De Roest et al. 2013; Whelan et al. 2018). These characteristics include low fermentability, insolubility and increased viscosity of the DF (Atzler et al. 2021a). Also, tailored application of DFs with different characteristics to different IBS-subtypes (e.g. predominantly diarrhoea or constipation) could be useful (So et al. 2021; Atzler et al. 2021b).

This study investigates the effect of fibre enrichment on techno-functional, rheological, sensory and predicted nutritional quality in a low FODMAP pasta (based on wheat starch and gluten). Therefore, pasta dough characteristics (water absorption, gluten aggregation) and cooked pasta quality (water optimal cooking time, cooking loss, colour, hardness, stickiness, tensile strength) were evaluated. Furthermore, the predicted nutritional quality (sugar release during in vitro starch digestion, total digestible starch content, FODMAP content) and sensory properties were determined. Results were compared to a low FODMAP control pasta (LFCP), not fortified with low-FODMAP fibre ingredients, and a commercial control pasta (CCP) which fulfils the low FODMAP criteria. The results of this study contribute significantly to future food design and can facilitate progress in personalised nutrition.

Methods and materials

Raw materials and chemicals

Prototypes with a low FODMAP content were prepared by replacing semolina with a mix of gluten-free wheat starch (Roquette, France) and vital gluten (Roquette, France) and by incorporating either bamboo fibre VITACEL BAF 200, powdered cellulose VITACEL L 600-30, psyllium VITACEL P95 (all supplied by J. Rettenmaier GmBH, Germany) and guar gum (Cargill, France). The CCP was purchased from Tesco (Dublin, Ireland), consisted of rice flour, white maize flour, yellow maize flour and an emulsifier (mono- and di-glycerides of fatty acids) and had a fibre content of 3.1 g/100 g according to the supplier. All chemicals were purchased from Sigma-Aldrich if not specified differently.

Compositional analysis

Products have been analysed regarding their composition including determination of protein content (based on a nitrogen-to-protein conversion factor of 5.7 for wheat-based products), moisture content (based on AACC method 44-15.02 (AACC International, 2000a)), ash content (based on AOAC 923.03 (AOAC International, 2005)) and fat content (utilising Soxhlet method and using SoxCap and SoxTec units (Foss UK Ltd., UK)) (AOAC International, 2005). Contents of resistant starch and digestible starch have been measured using the resistant starch assay kit (rapid) K-RAPRS supplied by Megazyme, Ireland. Total starch was calculated as the sum of digestible and resistant starch. Samples were freeze-dried and ground to a fine powder using a Tissue Lyser II (Hilden, Germany) before the starch determination. FODMAP contents have been analysed according to Ispiryan et al. (2019). For result based on the fresh weight of the pasta samples, the moisture contents of the products were taken into consideration. For the calculation of FODMAPs per serving, a serving size of 50 g of pasta was used.

Empirical dough analysis

LFCP (mixture of wheat starch and gluten in a ratio of 85% to 15%) and the low FODMAP fibre-enriched pasta samples (LFFP) were analysed for their water adsorption and gluten-aggregation properties. Mixtures of fibre-enriched pasta were prepared by replacing starch in the starch-gluten-mix. Recipes were calculated to reach a fibre content of 6 g/100 g in the cooked pasta aiming for a high in fibre claim according to EU regulation (Regulation (EC) No 1924/2006 2006). The addition levels of the single DF ingredients are demonstrated in Table 1.

Table 1. Recipe of the low FODMAP control pasta (LFCP) and pasta enriched with different dietary fibre ingredients.

	LFCP	Bamboo fibre	Cellulose	Psyllium	Guar gum
Starch (%)	85.00	71.00	74.00	64.00	67.00
Gluten (%)	15.00	15.00	15.00	15.00	15.00
Fibre Ingredient (%)	–	14.00	11.00	21.00	18.00
Table salt (%)	0.50	0.50	0.50	0.50	0.50
Tap water (%)	30.00	37.90	40.50	83.50	88.70

Notes: The quantities are given in % based on the sum of starch, gluten and fibre ingredients which results in 100%. Table salt and tap water are added additionally.

Farinograph

The water absorption was determined using the Farinograph-TS equipped with the FarinoAdd-S300 attachment (Brabender GmbH & Co. KG., Germany). Since pasta dough has a relatively low water addition level suitable for the extrusion process, the crumbly nature of the dough led to inconsistent high values (>1200 BU). Thus, a water titration to determine the water addition level in the recipes was not applicable. Instead, a linear regression and an extrapolation, based on measurements of the torque at water addition levels above 40%, were used for the calculation of the water absorption. The total volume of dry ingredient was 200 g based on a moisture content of 14%. The mixing chamber was tempered to 30 °C, and a constant mixing speed of 63 rpm was chosen. The measurement started with 1 min premixing, followed by water addition using the Aqua Inject (Brabender GmbH & Co. KG.), and the torque measurement time was set to 10 min. The target consistency was identified by measuring the torques values of the low FODMAP control recipe at four different concentrations (45%, 50%, 55% and 60%) and using a linear regression to calculate the torque at a water addition level of 30% (determined in pre-trials; data not shown). A target torque of 500 FU was determined, and water addition levels of fibre-fortified pasta recipes were calculated aiming for this torque value. Dough torques were measured at a range of 45–60% water addition for insoluble fibres and a range of 60–70% for soluble/viscous fibres. Linear regressions with a R² of 0.98 were used to calculate the water addition level which was given based on the starch and fibre ingredient.

Interaction with gluten network development

The GlutoPeak (Brabender GmbH & Co. KG., Germany) was used to investigate the interactions of DF ingredients with the gluten network. The measurement parameters were chosen as reported by Sahin et al. (2021), and the mixture of starch, vital gluten and the DF ingredient in the ratio reported in Table 1 were evaluated. The LFC and recipes including cellulose and bamboo fibre were conducted at a ratio of 56% dry ingredients and 44% tempered deionised water (36 °C). Recipes including psyllium and guar gum were measured at a ratio of 50/50 since the maximum torque (MT) exceeded the limit of 110 BU. The weight of solid samples was based on a moisture of 14%, and a total weight of 18 g was used. The impact of DF on gluten network development was evaluated by comparing the MT and the peak maximum time (PMT).

Pasta-making process

Pasta was prepared by mixing the dry ingredients in a Kenwood mixer bowl using a Titanium Major (KM020) mixer (Kenwood, Havant, UK) equipped with a K-beater for 2 min first to achieve a homogeneous mix. Table 1 illustrated the different recipes. Fibre-fortified pasta recipes were designed to reach a total fibre content of 6 g/100 g in the cooked pasta by replacing wheat starch by fibre ingredients. Concentrations were calculated considering the water absorption during cooking determined in pre-trials (data not shown). Following, tap water (30 °C) was added to the dry ingredients and mixing at minimum speed was performed. A water addition level of 30% was used to produce the LFCP. All other water amounts were adjusted using the Farinograph-TS. The extrusion process was performed as reported by Sahin et al. (2021). Briefly, the dough was placed in an PN 300 extruder (Häussler, Heiligkreuztal, Germany), fitted with a spaghetti die (internal diameter 2 mm), and strands of 20 cm length were cut with scissors. The fresh pasta was placed in a plastic food box to avoid moisture loss before cooking. The pasta production procedure was the same for all types of pasta.

Techno-functional properties of pasta quality

Optimal cooking time, cooking loss, tensile strength, hardness, stickiness and colour were determined to evaluate the quality of the pasta samples. The quality characteristics of the fibre-fortified pasta were compared to the LFCP and the CCP.

Optimal cooking time

Each sample was analysed regarding its optimal cooking time according to the method described by Hager et al. (2012). Samples were then cooked according to this time before they were further analysed (Hager et al. 2012).

Cooking loss

The content of dry matter in the cooking water was measured to determine the cooking loss. Measurements were carried out according to the method reported by Hager et al. (2012), which is based on the approved AACC method 66-50 (AACC International 2000b).

Texture properties of cooked pasta

Textural parameters including hardness, stickiness and tensile strength were analysed using a TA-XT2i texture analyser (Stable Micro Systems, Godalming, UK) with a 5 kg load cell, combined with the respective measuring probe. The measurements were conducted as reported by Sahin et al. (2021).

Colour of cooked pasta

The colour of the samples was measured using a colourimeter CR-400/410 (Konica Minolta Holdings Inc., Osaka, Japan). Before the measurement, the colourimeter was calibrated and subsequently, the colour was measured at 30 different locations on the surface of the pasta. After the measurement, the colour was evaluated using the L*-, a*- and b*-value of the CIE colour scheme. However, only L*-values were used as a*- and b*-values did not show significant differences (data not shown).

Microstructure

The microstructure of the cooked pasta samples was evaluated using Scanning Electron Microscopy (SEM) and confocal laser scanning microscopy (CLSM).

Images of the inner layer of the pasta were taken via SEM. The cooked pasta was cut into cross-sections and subsequently freeze-dried. Freeze-dried samples were fixated on stubs (G 306; 10 mm × 10 mm Diameter; Agar Scientific, UK) using carbon tape (G3357N; Carbon Tabs 9 mm; Agar Scientific, UK). Afterwards, samples were coated in a gold-palladium alloy (ratio of 80/20) using a Polaron E5150 sputter coating unit. Images were taken using a JEOL Scanning Electron Microscope (JSM-5510, Jeol Ltd., Tokyo, Japan) and applying a 5 kV voltage, a 20 mm working distance and a magnification factor of 1000.

CLSM imaging of the protein-carbohydrate structure of the samples was conducted using a confocal laser scanning microscope (Fluoview FV1000 incorporating an IX81 inverted microscope, Olympus, Hamburg, Germany) at 20x magnification. Both, a He-Ne laser (excitation wavelength 633 nm) and an Argon laser (excitation wavelength 488 nm), were used as samples, which were dyed with a combination of Nile blue A perchlorate and FITC. Samples were prepared for analysis as follows. Frozen samples were cut into cross-sections using a scalpel and subsequently incubated in the dark in chamber dishes for 30 min in 1-mL Nile blue A perchlorate (1% w/v solution in deionised water). After thorough rinsing with deionised water, the sample was dyed again for 10 s using a FITC solution (0.3% w/v in deionised water). Samples were again thoroughly rinsed with deionised water. For imaging, the dyed pasta cross-sections were placed in a chamber dish and weighted down with a cover slip to ensure contact of the sample surface and the surface of the well.

In vitro starch digestibility

In vitro starch digestibility was determined based on the degradation of digestible starch by α-amylase and the subsequent release of reducing sugars. An enzyme assay was carried out as previously described by Brennan and Tudorica (2008). The amount of reducing sugars released (RSR; expressed in % of digestible starch content (DSC)) was calculated using the following equation: $$\mathrm{\text{RSR}}\mathrm{ }\left(\mathrm{\%}\right)=\frac{A_{\mathrm{\text{Sample}}}\mathrm{*}500\mathrm{*}0.95}{A_{\mathrm{\text{Maltose}}}\mathrm{*}\mathit{\text{DSC}}}\mathrm{*}100$$ where A_sample represents the absorbance measured in the sample at 546 nm, A_Maltose represents the absorbance measured for a maltose solution (1 mg/mL; w/v) at 546 nm, DSC represents the digestible starch content (in mg), 500 mL is the total volume of buffer used and 0.95 is the conversion factor from maltose to starch.

Sensory analysis

Sensory evaluation of the pasta samples was conducted using quantitative descriptive analysis (QDA™). All panellists (n = 8; all female) had a minimum of 4 years sensory testing experience and additionally attended four 2-h training sessions on the assessment of pasta. The panel identified and agreed on a list of nine attributes (adapted from previous research) which discriminated across the samples (Table 2). Pasta samples were cooked for the determined optimal cooking time. Following cooking, nine strands of pasta were placed in a closed plastic container labelled with a random three-digit code and served to panellists according to a balanced design. Panellists evaluated the samples in triplicate over three 1-h sessions. Attributes were rated on a 10-cm line scale with appropriate end anchors. The overall quality of the appearance, flavour and texture of each sample was rated on a five-point scale where 1 = poor, 2 = fair, 3 = good, 4 = very good and 5 = excellent. Sensory testing was conducted under white light in a sensory testing facility (ISO 8589:2007) using Compusense Cloud Software (Compusense Inc., Ontario, Canada). Unsalted crackers and filtered water were used as palate cleansers.

Table 2. List of sensory attributes with agreed definitions and scale end anchors properties.

Attributes	Description	Scale anchors
Colour intensity	The intensity of colour from dark to bright	0 – Dark	10 – Bright
Surface smoothness	Glossy, watery and smooth surface	0 – Not smooth	10 – Very smooth
Firmness	The force required to bite down on three pasta strands using the molar teeth during the first bite	0 – Not firm	10 – Very firm
Chewiness	Force required to masticate the three pasta strands three times with the molars	0 – Not chewy	10 – Very chewy
Adhesiveness to teeth	Characterised by pasta pieces that adhere to the teeth during chewing	0 – Not adhesive	10 – very adhesive
Graininess	The coarseness /feeling of sand grains between the teeth just after swallowing	0 – Not grainy	10 – very grainy
Sweetness	The level of sweetness in the pasta measured during chewing.	0 – Not sweet	10 – Very sweet
Bitterness	The level of bitterness in the pasta measured during chewing	0 – Not bitter	10 – Very bitter
Aftertaste	Combination of roughness in flavour such as: astringency, bitterness and harshness perceived 30 s after swallowing.	0 – No aftertaste	10 – Very strong aftertaste

Statistical analysis

All measurements were performed in triplicate and as described. Statistical analysis was performed using SPSS version 28.01 (SPSS Inc., Chicago, IL) and Microsoft Excel version 2207 (Microsoft Redmond, WA) with the criterion for statistical significance set at α ≤ 0.05. A one-way analysis of variance (ANOVA) was used for the statistical evaluation and determination of significant differences among the physicochemical, technological and nutritional properties. For the sensory data, a two-way ANOVA was applied, considering samples, assessors and their interaction. Significant differences were determined using Tukey’s post hoc test, and the alpha value of 0.05 was used. Furthermore, a Pearson correlation analysis was carried out to correlate the technological data with the sensory data.

Results

Properties of dough and cooked pasta

Properties of pasta dough

The impact of DF ingredients on water absorption of the dough and gluten network development is illustrated in Table 3. The addition of insoluble (bamboo, cellulose) and soluble fibres (psyllium, guar gum) led to an increase in water absorption compared to the LFCP. However, soluble fibres caused a higher increase as absorption values above 80% were determined. Semolina-based pasta, as a comparison, has a water absorption value of 30% based on flour (Neylon et al. 2021; Sahin et al. 2021).

Table 3. Technological properties of dough formulations/cooked pasta and the compositional/nutritional properties (presented on an “as is” basis) of commercial control (CCP), low FODMAP control (LFCP) and the fibre-fortified products.

	CCP	LFCP	Bamboo fibre	Cellulose	Psyllium	Guar gum
Farinograph
Water absorption (%)	n.a.	30.00^1	37.90^a	40.50^a	83.50^b	88.70^b
GlutoPeak
MT (BU)	n.a.	9.33 ± 0.58*^a	94.00 ± 1.00*^c	71.33 ± 3.21*^b	108.00 ± 2.00**^d	143.33 ± 4.51**^e
PMT (s)	n.a.	11.33 ± 0.58*^a	26.67 ± 1.53*^b	52.00 ± 1.73*^c	11.00 ± 1.00**^a	9.00 ± 1.00**^a
Cooking parameters
OCT (min)	12.50^c	6.00^b	6.00^b	6.00^b	2.00^a	11.50^c
Cooking loss (%)	5.54 ± 0.86^bc	2.92 ± 0.02^a	7.79 ± 0.42^d	5.47 ± 1.30^bc	6.99 ± 1.34 ^cd	2.08 ± 0.24^a
Moisture (after cooking) (%)	49.13 ± 1.62^ab	51.02 ± 2.46^b	51.87 ± 1.88^b	48.18 ± 1.05^ab	70.98 ± 0.89^c	68.42 ± 0.99^c
Colour
L* (–)	69.41 ± 1.18^b	88.75 ± 6.21^c	87.53 ± 6.53^c	87.37 ± 6.79^c	52.20 ± 6.08^a	76.53 ± 6.61^b
TPA
Hardness (N)	1.95 ± 0.16^d	2.17 ± 0.24^e	1.05 ± 0.04^b	1.32 ± 0.11^c	n.d.^1	0.61 ± 0.08^a
Stickiness (N)	1.95 ± 0.39^a	4.84 ± 0.52^c	2.13 ± 0.09^a	3.91 ± 1.07^b	n.d.^1	4.41 ± 0.78^bc
Tensile Strength (N)	1.79 ± 0.31^b	2.36 ± 0.39^c	n.d.^1	1.71 ± 0.30^b	n.d.^1	1.37 ± 0.09^a
Ash (g/100 g)	0.15 ± 0.06^bcd	n.d.^2	0.06 ± 0.02^ab	0.09 ± 0.06^b	0.24 ± 0.01^d	0.12 ± 0.01^bc
Protein (g/100 g)	2.23 ± 0.29^a	3.73 ± 0.27^b	3.30 ± 0.18^b	3.28 ± 0.20^b	3.73 ± 0.35^b	2.73 ± 0.16^a
Fat (g/100 g)	0.47 ± 0.04^c	0.26 ± 0.03^ab	0.45 ± 0.11^c	0.38 ± 0.01^bc	0.55 ± 0.07^c	0.43 ± 0.06^c
Total dietary fibre (based on calculation (g/100 g) ***	1.70 ± 0.13	0.40 ± 0.08	6.10 ± 0.10	6.08 ± 0.05	6.11 ± 0.12	6.10 ± 0.10
Digestible Starch (g/100 g)	39.79 ± 5.41^d	36.14 ± 5.30 ^cd	26.94 ± 3.62^b	28.94 ± 1.22^bc	12.27 ± 0.70^a	15.25 ± 0.36^a
Total FODMAP (g/100 g)	n.d.^2	n.d.^2	n.d.^2	n.d.^2	n.d.^2	n.d.^2
Fructan (g/100 g)	n.d.^2	n.d.^2	n.d.^2	n.d.^2	n.d.^2	n.d.^2
GOS (g/100 g)	n.d.^2	n.d.^2	n.d.^2	n.d.^2	n.d.^2	n.d.^2
Polyol (g/100 g)	n.d.^2	n.d.^2	n.d.^2	n.d.^2	n.d.^2	n.d.^2

Notes: MT – Maximum torque; PMT – peak maximum time; OCT – Optimal cooking time; L* – Lightness.

Mean ± standard deviation with different letters in row are significantly different (p ≤ 0.05).

n.a. – not applicable; n.d.¹ – not determinable as values are below minimum force; n.d.² – not determinable as contents are below 0.01 g/100 g for Ash or 0.005 g/100 g for FODMAPs.

¹Determined in preliminary trials (data not shown).

*GlutoPeak trials were executed using a 50/50 ratio between solid and liquid.

**GlutoPeak trials were executed using a 56/44 ratio between solid and liquid.

***Content based on calculation: The fibre content was calculated considering the recipe and the fibre content of each ingredient used. Moreover, the water uptake during cooking was included in the calculation which causes a decrease in fibre content compared to the uncooked pasta.

The incorporation of DF significantly increased the MT compared to the LFCP. The addition of the soluble DF psyllium and guar gum resulted in MTs above 100 BU, while the insoluble fibres bamboo and cellulose caused MTs between 70 and 100 BU. However, a significant increase in the PMT was only observed for pasta recipes containing bamboo fibre or cellulose. Cellulose incorporation resulted in the highest PMT. The gluten network development characteristics of semolina pasta determined in previous studies showed a MT of 45 BU which is significantly lower than the fibre-rich pasta doughs, and a significantly longer development time (PMT = 91 s) (Neylon et al. 2021).

Properties of cooked pasta

The results for the technological properties of cooked pasta are shown in Table 3. Significant differences between LFCP and fibre-fortified low FODMAP pasta were determined for all parameters. The moisture content of cooked pasta reflects the water absorption during cooking. The levels ranged from 46% to 71%. The fortification with soluble DFs (psyllium, guar gum) resulted in significantly increased moisture contents around 70%, while the inclusion of cellulose caused the least water uptake during cooking (48%).

CCP showed the highest OCT amongst all pasta samples evaluated. The OCT of the LFCP and pasta containing bamboo fibre or cellulose was between 5.5 and 6.0 min. The fortification with guar gum led to a significantly higher OCT (>10 min). In contrast, the addition of psyllium significantly lowered the OCT to 2 min. As comparison, semolina pasta analysed in previous studies showed an OCT of 5.5 min, very similar to low FODMAP pasta fortified with insoluble fibre (Sahin et al. 2021).

The cooking loss of the different pasta samples ranged from 2% to 8%. CCP had a cooking loss similar to the pasta fortified with bamboo, cellulose and psyllium. The incorporation of bamboo, cellulose and psyllium significantly increased the cooking loss (7.79%, 5.47%, 6.99%, respectively) compared to the LFCP (2.92%). On the contrary, the fortification with guar gum caused a decreased in cooking loss (2.08%). Semolina pasta analysed in previous studies had a cooking loss of 5–6% (Neylon et al. 2021; Sahin et al. 2021).

The lightness of the samples was also found to be depending on DF used. As comparison, semolina pasta was showed to have a L*-value of about 70 (Sahin et al. 2021). The CCP had lightness values below 80, and only the guar gum containing pasta was measured to have a similar lightness value. The LFCP and pasta containing bamboo fibre or cellulose led to L*-values above 80. No significant differences between LFCP and the pasta fortified with insoluble DFs occurred, however, the addition of soluble DF decreased the lightness with psyllium causing the lowest L*-value of 52.

The effects of DF on texture were evaluated using the texture analyser and included measurements of hardness, stickiness and tensile strength of the pasta. The texture properties of psyllium containing pasta could not be determined due to the occurrence of strong gelling and an accelerated rate of breaking of pasta strands. Pasta-containing bamboo fibre, cellulose and guar gum had a significantly lower hardness (1.05 N, 1.32 N, 0.61 N, respectively) compared to the LFCP (2.17 N). The hardness of the CCP was comparable to the LFCP and significantly higher compared to pasta fortified with DF. Semolina pasta previously showed a hardness value of 2.2 N, which is not significantly different from LFCP (Sahin et al. 2021).

The incorporation of guar gum resulted in no significant change in stickiness compared to LFCP, while cellulose addition slightly reduced stickiness. CCP and low FODMAP pasta with bamboo fibre had significantly lower stickiness values (1.9–2.2 N).

The incorporation of all four DF reduced tensile strength, compared to the LCFP. Semolina pasta was previously measured and had a tensile strength of 0.3 N (Neylon et al. 2021). The tensile strength of the pasta fortified with bamboo fibre and psyllium was not measurable as the values were below the minimum detectable force of the measurement system. Furthermore, there was no significant difference in the tensile strength of the cellulose containing LFFP and the CCP (1.7 − 1.8 N).

Chemical and nutritional properties of the cooked pasta

A summary of the chemical and nutritional analysis is depicted in Table 3. Ash and fat contents were determined to be below 1 g/100 g and only small differences were observed. The protein contents ranged from 2.45% to 4.09%. Pasta fortified with guar gum and the CCP showed significant differences in protein content compared to the LFCP. The most significant differences in the composition of the samples were observed for the digestible starch. The digestible starch content in CCP (40 g/100 g) was similar to the LFCP (36 g/100 g). On the contrary, the DF fortification of the low FODMAP pasta led to a significant decrease in digestible starch due to wheat starch replacement. However, the degree to which the starch amount is lowered depends on whether insoluble or soluble DF were used. For insoluble DF (bamboo fibre, cellulose), digestible starch contents between 26 and 29 g/100 g were measured. Soluble-viscous DF (psyllium, guar gum) decreased the amounts of digestible starch values to 12% and 15%, respectively.

Microstructure

Both the analysis with CLSM and SEM (Figure 1) revealed significant differences in the microstructure. SEM images show that LFCP, and low FODMAP pasta containing bamboo fibre or cellulose contain gelatinised starch granules embedded in a protein network. However, the microstructure of the CCP and pasta fortified with psyllium or guar gum significantly varied. Both psyllium and guar gum containing pasta were found to have lower amounts of starch granules and a more compact structure illustrated by the limited number of holes. For psyllium, a very homogenous structure, which does not allow for differentiation between proteins and carbohydrates, was observed. The CCP has a microstructure consisting of a homogenous network with a very low number of starch granules and a high amount of larger holes.

Figure 1. Photographs [.1], CLSM [.2] and SEM [.3] images of commercial control pasta (CCP) [A], low FODMAP control pasta (LFCP) [B] and low FODMAP pasta fortified with bamboo fibre [C], cellulose [D], psyllium [E] and guar gum [F].

The microstructure observed in the CLSM images is in line with the microstructure seen in the SEM images. This technique additionally allowed to distinguish between starch (green), proteins (red) and fibres (red). All products except the CCP and the LFFP containing psyllium showed distinguished starch granules and a protein network (light red colour). CLSM pictures of pasta containing bamboo fibre, cellulose or guar gum showed the presence of fibres (more fibrous, edgy shapes of red structures). However, the fibre particles differed in shape and size. Micrographs of pasta fortified with bamboo fibre or guar gum showed longer and thin fibre particles with a length above 100 µm and an angular shape. Fibre particles with a higher diameter and a shorter length below 100 µm were observed in cellulose containing pasta. Furthermore, guar gum caused links between the various particles. CLSM images of the pasta enriched with psyllium showed no clear structure as neither starch granules nor protein clusters were identified. The CCP, in turn, showed a structure strongly disrupted by holes that were not dyed and, therefore, appeared black.

In vitro starch digestibility

The release curves of reducing sugars (RSR) over a period of 5 h from a simulated in vitro starch digestion are presented in Figure 2. It was found that the in vitro starch digestibility of the pasta enriched with bamboo fibre, psyllium and guar gum was lower compared to the LFCP. Fortification with psyllium led to a reduction of the final RSR content to 12%. The addition of bamboo fibre and guar gum resulted in final RSR contents of 8% and 9%. No significant differences were seen in the starch digestibility of cellulose containing pasta compared to the LFCP (final RSR > 15%). The CCP showed a final RSR content of 10%.

Figure 2. Release of reducing sugars (RSR) in percent during in vitro starch digestion based on the digestible starch content of commercial control pasta (CCP), low FODMAP control pasta (LFCP) and low FODMAP pasta fortified with bamboo fibre, cellulose, psyllium and guar gum.

Sensory analysis

Sensory characteristics

The sensory profiles of the different pastas were determined by a trained panel, which established and evaluated nine attributes describing the pastas’ appearance, texture, flavour and the overall quality of appearance, texture and flavour. The results of sensory analysis are summarised in Table 4.

Table 4. QDA mean panel data and post hoc test groupings for the sensory properties of commercial control pasta (CCP), low FODMAP control (LFCP) and the fibre-fortified samples.

	CCP	LFCP	Bamboo	Cellulose	Psyllium
Appearance
Colour intensity	7.64 ± 1.18^a	7.30 ± 1.44^a	7.17 ± 1.56^a	6.68 ± 1.55^b	0.88 ± 0.53^c
Surface smoothness	7.03 ± 1.63^a	6.34 ± 1.53^ab	5.65 ± 1.89^b	5.25 ± 1.65^b	2.40 ± 1.19^c
Texture
Firmness	5.30 ± 1.56^a	2.92 ± 1.73^b	2.51 ± 1.33^b	2.50 ± 1.40^b	6.04 ± 1.51^a
Chewiness	4.28 ± 1.71^bc	3.33 ± 1.39^cd	2.56 ± 1.41^d	2.59 ± 1.04^d	5.60 ± 1.15^ab
Adhesiveness	2.30 ± 1.73^c	3.90 ± 2.11^bc	4.23 ± 1.99^ab	4.34 ± 1.79^a	3.74 ± 1.81^bc
Graininess	3.17 ± 2.70^a	1.86 ± 1.54^b	3.03 ± 1.89^a	3.58 ± 2.34^a	3.72 ± 1.66^a
Flavour
Sweetness	1.65 ± 1.29^a	2.16 ± 1.55^a	1.53 ± 1.41^a	1.77 ± 1.52^a	1.92 ± 1.33^a
Bitterness	0.45 ± 0.77^c	0.40 ± 0.55^c	0.68 ± 0.93^b	0.80 ± 1.06^ab	0.49 ± 0.78^c
Aftertaste	0.95 ± 1.58^a	1.45 ± 1.93^a	1.22 ± 1.54^a	1.87 ± 1.96^a	0.93 ± 1.47^a
Overall quality
Quality of appearance	3.67 ± 0.87^a	2.46 ± 0.98^a	1.92 ± 0.65^b	1.71 ± 0.75^b	1.54 ± 0.51^b
Quality of texture	3.29 ± 0.95^a	2.46 ± 1.02^a	2.25 ± 0.61^ab	1.96 ± 0.75^b	2.83 ± 0.70^a
Quality of flavour	3.42 ± 0.93^a	2.33 ± 1.24^b	2.17 ± 0.70^b	1.71 ± 0.62^b	1.75 ± 0.61^b

Note: Mean ± standard deviation with different letters in column are significantly different (p ≤ 0.05).

Addition of guar gum significantly lowered the colour intensity of pasta in comparison to all other samples assessed. The psyllium pasta also resulted in significantly lower colour intensity ratings in comparison to CCP, LFCP and bamboo pasta samples. The surface smoothness of the fibre-fortified pasta was lower compared to LFCP and CCP, while guar gum and psyllium pastas were rated to have the least smooth surface.

The firmness and chewiness of the fibre-enriched pasta with bamboo fibre, cellulose or psyllium compared well to the LFCP, while the pasta with guar gum addition and the CCP were perceived as firmer and chewier. Pasta fortified with bamboo fibre, cellulose or psyllium were perceived to have the same adhesiveness as the LFCP; the guar gum containing pasta was rated to be the most adhesive, and the CCP, in turn, the least. All fibre-enriched pasta samples had a slight trend towards more grainy perception of the texture compared to the LFCP, although only the ratings for cellulose and guar gum pasta were significantly higher. The graininess of the CCP was similar to fibre-fortified pasta. Overall, pasta containing bamboo fibre, cellulose or psyllium were rated to have a similar texture and pasta with guar gum was perceived as firmer and chewier. The flavour characteristics of all pasta samples were very similar. No significant differences in sweetness, bitterness and aftertaste were found across the pasta samples. The overall quality of appearance, texture and flavour of the low FODMAP pasta samples fortified with DF were rated to be similar to the LFCP; the appearance of the fibre-enriched pasta received slightly lower ratings, the texture of the guar gum pasta was perceived to be more pleasant and the flavour of the psyllium pasta received slightly lower ratings. No other significant differences were seen in the overall quality of the fibre-enriched pastas and the LFCP. The CCP received was rated the highest in the overall quality.

Correlation of sensory and technological characteristics

A Pearson correlation was executed to assess interactions and correlations between the technological parameters and sensory properties describing the appearance and texture. The results are illustrated in Table 5. Only strong positive or negative correlations, which were significant (p < 0.05; r > 0.5) were relevant.

Table 5. Results of the Pearson correlation analysis for the technological quality parameters (hardness, stickiness, tensile strength and colour) and the sensory properties (firmness, chewiness, adhesiveness, graininess, quality of appearance and quality of texture).

	L*	Hardness	Stickiness	TS	Surface smoothness	FAB	C	A	G	QA	QT
L*	1.000	−0.177	−0.091	−0.122	−0.352	0.048	0.046	−0.023	−0.158	−0.111	−0.136
Hardness	−0.177	1.000	0.430	0.710**	0.397	0.297	0.150	−0.289	−0.164	0.626**	0.517**
Stickiness	−0.091	0.430	1.000	0.076	0.136	0.010	0.003	−0.149	−0.118	0.179	0.085
TS	−0.122	0.710**	0.076	1.000	0.389	0.019	−0.012	−0.166	−0.155	0.471	0.379
Colour intensity	−0.162	0.340	0.079	0.490	0.598**	−0.437	−0.527**	−0.148	−0.188	0.323	0.218
Surface smoothness	−0.352	0.397	0.136	0.389	1.000	−0.083	−0.207	−0.374	0.045	0.366	0.382
FAB	0.048	0.297	0.010	0.019	−0.083	1.000	0.754**	−0.232	0.211	0.366	0.383
C	0.046	0.150	0.003	−0.012	−0.207	0.754**	1.00	0.038	0.245	0.189	0.195
A	−0.023	−0.289	−0.149	−0.166	−0.374	−0.232	0.038	1.000	0.313	−0.420	−0.515**
G	−0.158	−0.164	−0.118	−0.155	0.045	0.211	0.245	0.313	1.00	−0.231	−0.231
QA	−0.111	0.626**	0.179	0.471	0.366	0.366	0.189	−0.420	−0.231	1.000	0.813**
QT	−0.136	0.517**	0.085	0.379	0.382	0.383	0.195	−0.515**	−0.231	0.813**	1.000

Note: TS – Tensile strength; FAB – Firmness after first bite; C – Chewiness; A – Adhesiveness; G – Graininess; QA – Quality of Appearance; QF – Quality of Flavour; QT – Quality of texture.

*Significant at p < 0.05 and **strong positive or negative correlation with r > 0.5.

It was found that only the hardness of the pasta correlated with the sensory attributes. Hardness had a positive correlation with the quality of appearance (r = 0.626) and the quality of texture (r = 0.517).

Several correlations between the sensory attributes as well as texture characteristics were found. The firmness of the pasta positively correlated with the tensile strength. For the quality of appearance, a positive correlation with the quality of texture (r = 0.813) occurred. Furthermore, the quality of texture also negatively correlated with the adhesiveness (r = −0.515).

Discussion

The incorporation of the different fibre ingredients in low FODMAP pasta resulted in different technological characteristics of the doughs and the cooked pastas and sensory attributes. These differences can be substantially attributed to the water holding capacity of the DF and the increase in viscosity caused by the incorporation of the fibre ingredients (Neylon et al. 2021; Sahin et al. 2021).

The results of the farinograph highlight the significance of the water holding capacity of the various DF ingredients since an increase in the water absorption level was observed. The increase of water absorption levels by DF is a commonly reported effect. However, soluble dietary fibre showed a higher water absorption compared to insoluble fibre. Indeed, this discrepancy can be explained with the different water holding capacities of these DF. Both psyllium and guar gum were previously reported to have significantly higher water holding capacities (20.0 g and 21.5 g H₂O/g solid, respectively) than bamboo fibre and cellulose (8.3 g and 4.0 g H₂O/g solid, respectively) (Atzler et al. 2021b), which led to the necessary adjustment of solid/liquid ration in the GlutoPeak measurement.

Furthermore, the competition for water between the fibre and protein, mainly glutenin/gliadin, putatively caused the increase in MT and PMT (Sahin et al. 2021). Moreover, the increase of the MT by bamboo fibre and cellulose can be explained by the ability of these DF to act as plasticisers (Nawrocka et al. 2017).

Although the water holding capacity can be seen as the main factor for these effects, two other parameters impacting the dough properties were identified. Firstly, the particle size of insoluble DF, can have a significant influence on technological dough and product characteristics. The bamboo fibre had a significantly larger particle size than cellulose (350 μm versus 30 µm; as specified by the supplier). Therefore, disruption of homogeneity by fibre particles, dependent on particle size, can be seen as a possible cause for the differences in gluten network development time and strength (Zhang and Moore, 1997; Tudorica et al. 2002; Sangnark and Noomhorm 2003; Brennan and Tudorica 2007). Additionally, considering an increase in viscosity caused by the formation of a gel network is characteristic for soluble, viscous DF. The gel network formation could have caused the high MT and the low PMT values. This can be explained as a rise in viscosity increases the force needed for the kneading of the dough (Atzler et al. 2021b).

Differences in the water holding capacities of the different fibre ingredients were decisive for the results of the moisture after cooking, the hardness, the stickiness and the tensile strength. A much higher moisture after cooking was observed for both psyllium and guar gum, which is related to their high water holding capacities (Horstmann et al. 2018). On the other hand, the reduced pasta hardness and stickiness caused by the addition of bamboo fibre, cellulose or guar gum was putatively caused by the competition for water between the DFs and the starch-protein network (Tudorica et al. 2002; Brennan and Tudorica 2007). The reduction of freely available water results in lower levels of starch gelatinisation and gluten network development. Bamboo fibre and cellulose also reduced stickiness caused by the same effects (Tudorica et al. 2002; Brennan and Tudorica 2007). The significance of the effects of the different water holding capacities is further highlighted by the fact that the rate at which hardness, stickiness and tensile strength are reduced aligns with the increased water holding capacity specified by the supplier for these DFs.

Furthermore, difference in the pasta characteristics can also be attributed to differences in particle size and gel-network strength of the different fibre ingredients. Cooking loss and the OCT are primarily influenced by these two DF properties. The disruption of the outer protein layer, which is responsible for the retention of organic material, was observed in the pasta with bamboo fibre addition; incorporating bamboo fibre led to a significant increase in cooking loss. This could be explained by the elongated shape of fibre particles, which were also visualised in the CLSM images of the bamboo fibre containing pasta (Gelencsér et al. 2008; Krishnan and Prabhasankar 2012; Li et al. 2014).

The incorporation of psyllium resulted in a significantly reduced cooking time and increased cooking loss due to the lack of formation of either a defined protein-starch network or a gel network. However, guar gum had the opposite effect due to the formation of a stronger gel network reflected by high MT values. Visual evaluation of the microstructure seen in the SEM and CLSM images supports these findings. In the CLSM images, no distinct starch granules embedded in a protein network, but instead, diffuse and undefined starch-, fibre- and protein structures were visible. The absence of any defined network allows water to quickly enter the system and amylose to leach into the cooking water. It was frequently reported that the amount of cooking loss is related to the strength of the protein network on the outer layer of the pasta (Gelencsér et al. 2008). The effects of psyllium regarding network formation caused a general decrease in pasta quality reflected by the technological quality parameters (hardness, stickiness and tensile strength) being not measurable. Guar gum was the only DF ingredient that reduced the cooking loss and increased the OCT, which can be explained by the gel network encapsulating the starch and the protein (Brennan and Tudorica 2007; Atzler et al. 2021b). This also aligns with the findings obtained for the microstructure using SEM and CLSM since the formation of a continuous network was observed in both types of micrographs. Furthermore, the higher MT values measured suggest a stronger gel network than the pasta containing psyllium. Although an increase in OCT compared to the LFCP was observed, the OCT of guar gum containing pasta was still in a reasonable range and compared well to the CCP.

Overall, all pasta except psyllium had an acceptable technological quality compared to the CCP. The values for moisture after cooking, cooking loss and lightness of fibre-fortified pasta samples were not significantly different from those determined for the CCP. Moreover, the selected DF ingredients did not impact sensory characteristic. This highlights the potential of these fibres to produce a pasta which is high in fibre and has a desirable quality. The results of the sensory trial support the findings made for the technological quality parameters of the pasta. This is especially seen in the fact that the hardness was found to have a significant correlation with the sensory quality.

A beneficial effect on the nutritional composition of the fibre-enriched pasta samples can be predicted based on the results of the in vitro starch digestibility. Generally, the replacement of wheat starch by DF caused a decrease in sugar release over time. The only exception was the pasta fortified with cellulose, as no significant differences were determined compared to the LFCP. Also, the CCP showed a lower RSR during digestion compared to the LFCP. The CCP was specified to have a dietary fibre content of 3 g/100 g. which explains RSR values and starch digestibility, similar to the fibre-fortified pasta (Foschia et al. 2015). The lower RSR during digestion can be explained by the partial replacement of easily digestible wheat starch with indigestible fibre ingredients (Tudorica et al. 2002; Brennan and Tudorica 2007; Foschia et al. 2015). Moreover, the different inclusion levels need to be considered. For psyllium and guar gum, higher addition levels were necessary as they significantly increased water absorption and adjustments were needed to reach 6 g fibre per 100 g cooked pasta (Sahin et al. 2021). A decreased RSR can also be attributed to the fibres’ water holding capacities or their abilities to increase viscosity (Tudorica et al. 2002; Brennan and Tudorica 2007; Foschia et al. 2015). An impairment of the starch digestibility following the addition of DFs is a well-known fact and has been observed for various food products, including bread and pasta (Tudorica et al. 2002; Brennan and Tudorica 2007; Sasaki and Kohyama 2012; Foschia et al. 2015). The addition of DF often leads to a decrease in starch digestibility by restricting the gelatinisation of starch. Gelatinisation of starch improves and allows starch digestion in the human gastrointestinal tract (Chung et al. 2006). The level at which starch digestibility was affected depended on the water holding capacity and the increase in viscosity. In fact, cellulose with the lowest water holding capacity (4 g H₂O/g solid in contrast to 8–20 g H₂O/g solid; specified by the suppliers) did not significantly impact starch digestibility. Similar to the effects on technological properties, this is caused by differences in the water holding capacity and particle size. The higher water holding capacity of bamboo fibre (8.3 g H₂O/g solid) leads to higher competition for water and, therefore, a decrease in starch gelatinisation. The decrease in gelatinised starch was also observed during SEM. The two soluble, viscous DFs had even higher water holding capacities (∼20 g H₂O/g solid) and possess the ability to form gel networks. This would explain that guar gum addition resulted in the highest impairment of starch digestibility. An increase in viscosity causes a decline in starch digestibility by reducing the accessibility of starch granules for hydrolysis via α-amylase (Cameron-Smith et al. 1994). Additionally, food structure has a significant impact on accessibility of enzymes for their specific substrate, in this case starch (Sahin et al. 2021). The more compact pasta containing guar gum or psyllium and the gel network formed covering the starch granules embedded in the protein matrix restricted the amylase to cleave the alpha (1–4) glycosidic bonds releasing reducing sugars.

Ultimately, only low FODMAP ingredients were selected for the formulations (Ispiryan et al. 2020); this could be analytically confirmed with no FODMAPs detected in any of the pasta samples. Conventional pasta with a higher fibre content is usually made from wholemeal flour or cereal-based fibres, such as wheat bran and is, therefore, found to be high in FODMAPs (Biesiekierski et al. 2011; Varney et al. 2017). Hence, the fibre-enriched pasta samples in this study are not only suitable for the low FODMAP diet given their low FODMAP contents but could also positively assist the diet being a good source of DF (Varney et al. 2017).

Conclusion

In summary, this study shows the impact of different low FODMAP dietary fibre ingredients on a low FODMAP pasta model system based on wheat starch and vital gluten. The results are impactful for further product development providing consumers suffering from IBS symptoms adequate nutrition. Reaching an adequate fibre supply in a low FODMAP diet can be challenging. All DF ingredients used in this study, except for psyllium, had an acceptable quality regarding their technological properties. Additionally, the fibre-enriched pasta performed similarly to the LFCP regarding the sensory qualities, suggesting that the fortification with DF does not negatively impact these properties. Moreover, three main properties – which can be used for predicting the impact of the DF on pasta quality – were identified. These three properties are the water holding capacity of both soluble and insoluble DF, the insoluble fibres’ particle size and the soluble fibres’ viscosity. This knowledge can be used for further research, studying the synergistic effect of the selected fibre ingredients in coexistence. The suggested DF ingredient, based on the outcome of this study, are the insoluble DF cellulose and the soluble DF guar gum.

Ethical approval

Sensory evaluation of pasta was conducted in accordance with the Institute of Food Science and Technology (IFST) Guidelines for Ethical and Professional Practices for the Sensory Analysis of Foods (IFST 2020)and informed consent was obtained from all sensory panellists prior to commencing the study.

Acknowledgement

The authors would like to thank Tom Hannon for technical support and Dr. Andrea Hoehnel for her support and the inspiring discussions.

Disclosure statement

No financial or non-financial competing interests.

Additional information

Funding

This work was funded by the Irish Department of Agriculture, Food and the Marine. Project Acronym: TALENTFOOD – 15F602.