Objective: To ascertain (1) whether the taste characteristics of a conventionally-salted (150 mM NaCl) soup can be reproduced in soups of substantially lower NaCl level with the help of added glutamate, and (2) whether calcium diglutamate (CDG) is equivalent to monosodium glutamate (MSG) in its effect on the taste of soup.
Design: Cross-sectional, with multiple measurements on each subject.
Setting: Healthy university students.
Subjects: A total of 107 volunteers, recruited by on-campus advertising.
Methods: Subjects tasted 32 soups, with all possible combinations of four NaCl concentrations (0–150 mM), four glutamate levels (0–43 mM), and two glutamate types (MSG, CDG).
Main outcome measures: Ratings of each soup on six scales (liking, flavour-intensity, familiarity, naturalness of taste, richness of taste, saltiness).
Results: A 50 or 85 mM NaCl soup with added CDG or MSG is rated as high as, or higher than, a 150 mM NaCl soup free of added glutamate on five of the six scales (the exception being saltiness). CDG and MSG have equivalent effects.
Conclusions: Addition of glutamate allows substantial reductions in Na content of soup, without significant deterioration of taste. CDG and MSG have equivalent effects, but use of CDG permits a greater reduction in Na intake.
Sponsorship: We acknowledge the financial support of the International Glutamate Technical Committee (IGTC).
High sodium intakes contribute significantly to the development of hypertension and its sequelae. Short-term trials (Sacks et al, 2001) show clearly that reducing Na intakes significantly lowers blood pressure. In the longer term, Law (2000) recently estimated that reducing sodium intakes by 100 mmol/day would decrease stroke mortality by 22% and ischaemic heart disease by 16% in Western societies. The vast majority of Na intake derives from salt added to processed foods during commercial production (James et al, 1987; Mattes & Donnelly, 1991). Hence, widespread availability and use of low-sodium equivalents for these high-salt processed foods would be an important factor facilitating reduction of community Na intake.
Soup is widely consumed, and usually very salty. Most Australian commercial soups have sodium levels between about 3.5 and 7.0 g/l, or approximately 150–300 mmol per litre (English & Lewis, 1991), levels very similar to those in other Western countries (US Department of Agriculture, Agricultural Research Service, 1999; Holland et al, 1991; Favier et al, 1995; van der Heijden et al, 1996) A soup meeting the Australian legal requirement for a ‘low-sodium’ food (no more than 1.2 g Na/kg) would, by contrast, provide only 52 mmol/l. Developing low-sodium soups with accep-table taste characteristics has, however, proved difficult (Claiborne, 1980).
Monosodium glutamate (MSG) has long been used in several Asian cuisines for its flavour, known as ‘umami’ in Japanese, and flavour-enhancement properties (Yamaguchi & Ninomiya, 2000). It has been suggested therefore that MSG could be used as a means of ‘orienting food selection toward a healthy diet composition’ (Bellisle, 1998). A handful of experimental studies have indicated that MSG enhances the flavour of soups of various NaCl concentrations (Okiyama & Beauchamp, 1998; Roininen et al, 1996; Yamaguchi & Takahashi, 1984). In principle, therefore, use of MSG could maintain flavour characteristics at lower NaCl concentration, but its disadvantage is that it contains sodium itself: MSG concentrations of the order of 2.0–4.0 g/l (as used by Yamaguchi & Takahashi, 1984; Okiyama & Beauchamp, 1998) increase total Na concentration by approximately 10–20 mmol/l, thereby partially negating the reduction in NaCl that MSG allows.
Calcium diglutamate (CDG, sometimes known as calcium glutamate), is also approved as a food additive by regulatory authorities. Being free of sodium, it could potentially achieve equivalent soup flavour characteristics at a lower total Na concentration. Its Ca content might also modestly enhance the consumer's total Ca intake. However, no previous published studies appear to have reported the effect of CDG on the taste characteristics of soups (or other foods), and it appears to be little used as an additive.
Our primary aim was to establish whether the taste characteristics of a conventionally salted soup could be reproduced in soups of substantially lower NaCl level with the help of added glutamate. We also sought to establish whether CDG was equivalent to MSG in its effects on taste characteristics. Our reference (‘conventional’) soup had no added glutamate and 150 mM NaCl: this NaCl level was chosen to be in the typical range for Australian soups, and also corresponds very closely to the mean preferred NaCl levels in soup in the studies of Yamaguchi and Takahashi (1984) and Roininen et al (1996).
Materials and protocol
The 120 participants were young adults, almost all of them university students enrolled in psychology courses. Average age was 22 y and 67% were female. Heavy smokers and persons who reported ever having suffered a reaction to MSG (whether confirmed or otherwise) were excluded. We did not ask about their soup-eating habits, but unpublished data from a recent Hobart dietary survey by the authors (Beard et al, 1997) indicate that only about 30% of local adults consume soup more than once a week. The protocol for the study was approved by the University of Tasmania Human Ethics Committee, and all subjects gave informed consent in writing.
We tested 32 soups based on a simple pumpkin soup, with no added salt or glutamates. This was produced by the Australian Defence Nutrition Research Centre at Scottsdale, Tasmania. It contained 160 g pumpkin and 60 g onion, cooked and blended up in 1 l of water, and then freeze-dried. The freeze-dried powder–after addition of NaCl and glutamates, as specified in the next paragraph–was re-constituted with water, 1000 ml per 60 g powder. (The glutamate content of the freeze-dried powder was 1.030 g/kg (IGTC, private communication). Hence, the endogenous glutamate level in the reconstituted soup was 0.062 g/l, ie, 0.3 mM. Glutamate values quoted in Results refer to added glutamate and exclude this small amount of endogenous glutamate.)
To different aliquots of this common freeze-dried starting material, NaCl and MSG were added to produce a series of 16 soups that included all possible combinations of four concentrations of added NaCl (after reconstitution, 0.0, 2.9, 5.0 and 8.8 g/l, ie. 0, 50, 85 and 150 mM Na) and four concentrations of added MSG (after reconstitution, 0.0, 2.0, 4.0 and 8.0 g/l, ie, 0, 11, 21 and 43 mM with respect to glutamate). A further 16-soup series was produced, using all possible combinations of the same four concentrations of added NaCl and four concentrations of added CDG (after reconstitution, 0.0, 2.2, 4.4 and 8.8 g/l, ie 0, 11, 21 and 43 mM with respect to glutamate). It may be noted that the four soups containing zero MSG were therefore identical to the respective four soups containing zero CDG.
Monosodium glutamate monohydrate and calcium di-glutamate tetrahydrate were provided by the Ajinomoto Corporation of Japan.
Over two consecutive days, all subjects tasted all 32 soups, maintained at 85–90°C. On one day, the 16 soups of the CDG series were sampled and on the other day the MSG series was sampled, with half of the subjects sampling the CDG series first. A 30 ml aliquot of each soup was tasted and swallowed after tasting. Sampling was carried out with salt and glutamate content withheld from subjects, and giving each person one exposure to each of the 16 soups in an individually generated random order.
Rating of soups
Participants were required to rate each soup for six taste characteristics. These reflected potential barriers to the acceptance of new soups: that the new soup had no taste (addressed by the ‘flavour intensity’ rating); that the taste was unsatisfying (‘richness of taste’) or artificial/chemical (‘naturalness of taste’) or different from normal soups (‘familiarity’) or lacked the salty flavour characteristic of soups (‘saltiness’). In addition, there was a rating for global palatability (‘liking’).
The scales are described below, in the order in which they appeared on the rating sheet.
The question was ‘How much do you like this soup sample?’ and the response options provided were ‘dislike extremely’ (later coded as 1 for data analysis), ‘dislike very much’ (2), ‘dislike moderately’ (3), ‘dislike slightly’ (4), ‘neither like nor dislike’ (5), ‘like slightly’ (6), ‘like moderately’ (7), ‘like very much’ (8), ‘like extremely’ (9). The numerical codes were not displayed on the printed scale.
The question was ‘How would you rate the flavour intensity of the soup?’ and the response options were ‘no flavour’ (1), ‘slight flavour’ (2), ‘moderate flavour’ (3), ‘very strong flavour’ (4), ‘extremely strong flavour’ (5).
The question was ‘How familiar is this soup to you?’ and the response options were ‘not at all familiar’ (1), ‘slightly familiar’ (2), ‘moderately familiar’ (3), ‘very familiar’ (4), ‘extremely familiar’ (5).
(d) Naturalness of taste
The question was ‘How natural does this soup taste to you?’ and the response options were as in (c), except that ‘familiar’ was replaced by ‘natural’.
The question was ‘How salty does this soup taste to you?’ and the response options were as in (c), except that ‘familiar’ was replaced by ‘salty’.
(f) Richness of taste
The question was ‘How rich does this soup taste to you?’ and the response options were ‘not at all rich’ (1), ‘slightly rich taste’ (2), ‘moderately rich taste’ (3), ‘very rich taste’ (4), ‘extremely rich taste’ (5).
Of the 120 subjects, 13 subjects were excluded from data analysis as they had omitted ratings for five soups or more; analyses are therefore based on 107 subjects. All analyses used the Statistica for Windows Version 5 software pack-age, and only comparisons with P<0.01 were treated as significant.
The data were subjected to three-way analysis of variance (ANOVA) with repeated measures, on the following factors: 4 (salt level)×4 (glutamate level)×2 (cation). (As noted above, the four zero-glutamate soups from the MSG series were identical to the soups of equivalent NaCl level from the CDG series. This, however, does not violate any statistical assumptions required for ANOVA, and prior 4 (salt level)×2 (cation) ANOVAs including only these eight soups confirmed that there were no significant main effects or interactions involving cation within this subset.) Further pair-wise comparisons were made using the Tukey HSD procedure.
For all six rating scales, ANOVA indicated very strong main effects for NaCl concentration (P<0.0001). Glutamate also showed a very strong (P<0.0001) main effect for five of the scales, its effect on naturalness of taste being marginally less so (P=0.0003). There were also very strong interactions between these two factors on all scales: for liking, naturalness of taste and saltiness, P<0.0001; for richness of taste, P=0.0003; for flavour-intensity and familiarity of taste, P=0.002. However, the main effect for cation (Ca vs Na) was non-significant for all scales, as were all other interaction effects.
Because the main effect for cation was not significant, and there were no significant interactions involving cation, data presented below are based on the average value of corresponding MSG and CDG soups. Table 1 identifies for each rating scale those soups whose mean ratings are statistically equivalent to, or significantly greater than, those of the reference soup (150 mM NaCl, 0 mM glutamate).
Table 1(A) shows the mean liking ratings, with corresponding values for MSG and CDG averaged. Our reference soup (150 mM NaCl, with zero glutamate) rated approximately mid-range, with a mean of 4.3 (s.d. 2.1). Of soups with NaCl below 150 mM, those with 85 mM NaCl plus glutamate rated significantly better than the reference soup, as did soups with 50 mM NaCl plus glutamate.
For flavour-intensity (Table 1(B)), the reference soup rated just above mid-range, with mean 3.1 and s.d. 1.0. Of soups with NaCl below 150 mM, all those containing at least 85 mM NaCl plus glutamate rated equivalently to or higher than the reference soup, as did the soup with 50 mM NaCl plus 43 mM glutamate.
The reference soup rated approximately mid-range, with mean 2.2 and s.d. 1.0, for familiarity (Table 1(C)). Soups with 85 mM NaCl plus glutamate rated significantly higher than the reference soup, while soups with 50 mM NaCl plus glutamate rated equivalently or significantly higher.
For naturalness of taste (Table 1(D)), the reference soup rated somewhat below mid-range, with mean 2.0 and s.d. 1.0. All soups with 85 mM NaCl or 50 mM NaCl rated significantly higher, and those with zero NaCl equivalent to this soup, regardless of glutamate content.
As shown in Table 1(E), the reference soup rated at mid-range on richness of taste, with mean 2.6 and s.d. 1.1. All glutamate-enhanced soups with 85 mM NaCl or 50 mM NaCl rated equivalently.
The saltiness (Table 1(F)) of the reference soup rated substantially above mid-range, with mean 3.4 and s.d. 1.2. No soup with lower NaCl level, either with or without glutamate, rated equivalently or significantly higher.
CDG and MSG have equivalent effects on taste characteristics
Our data indicate that the flavour-enhancing characteristics of glutamate in soup depend on the concentration of glutamate. The accompanying cation does not seem to have much impact. This is a new finding; direct comparisons of the taste effects of MSG and CDG interactions with NaCl have not been previously reported.
As discussed below, there may be some health advantage in replacing MSG by CDG. As both CDG and MSG are approved food additives, such a substitution should not encounter any significant regulatory obstacles, although the greater cost of CDG may be seen as a disincentive for the replacement.
Adding glutamate allows lower-salt soups with equivalent taste
Lowering salt content had a negative effect on the soups' taste ratings. Compared to our reference soup, our glutamate-free lower-salt soups rated relatively poorly: the 0 mM NaCl soup rated at least as good as the reference soup on only one of the six scales, and the 50 mM and 85 mM NaCl soups on only three.
However, addition of glutamate gave better results for two NaCl levels. The 50 mM NaCl soups with MSG/CDG were judged to be at least as good as the reference soup on four scales, or five when 43 mM glutamate was used. The 85 mM NaCl soups with MSG/CDG were considered at least as good as the reference soup on five of the six scales. All of these lower-salt soups were perceived as less salty than the reference soup.
Thus, apart from tasting less salty, an 85 mM NaCl soup with added glutamate–or a 50 mM NaCl soup with 43 mM glutamate–is equivalent in taste characteristics to our reference soup. It is uncertain how crucial perceived saltiness is in people's response to soup, but the observation that all the 50 mM and 85 mM NaCl soups were in fact liked better than the reference soup suggests that saltiness may not be a key determinant of liking for a soup.
Our results are broadly consistent with those of other researchers (eg Okiyama & Beauchamp, 1998; Roininen et al, 1996; Yamaguchi & Takahashi, 1984) who have examined the interaction of NaCl and MSG in soups, but their diverse experimental designs make detailed comparisons impossible. However, their data suggest that our results are broadly generalisable to several kinds of soup. Assessing whether they can be generalised to other types of food products will require further empirical investigation.
Implications of such soups for overall Na intake
Replacing a conventionally salted soup with a glutamate-fortified lower-salt soup could substantially lower the total daily Na intake of a soup consumer. The reduction in Na would be greater if CDG was the source of the glutamate. On an average day, 12% of Australians consume soup, the median serving being 383 g, with only minor gender differences (Australian Bureau of Statistics, 1999).
A typical serving of the reference soup (150 mM NaCl) would provide 57 mmol Na, compared to 33 for a serving of an 85 mM NaCl soup with CDG. Thus, with only a minor impact on hedonic response, this simple substitution would decrease the day's total Na intake by 24 mmol–not a trivial reduction, given that Australian males consume an average of 170 mmol/day and females 112 mmol/day (Beard et al, 1997). A 50 mM NaCl soup with 43 mM CDG would allow a reduction of 38 mmol/day in total Na intake, an even more impressive decrease.
The ultimate goal is a substantial reduction in community Na intake, at least to the upper limit of the recommended range (100 mmol per day; National Health and Medical Research Council, 1991), although the recent study by Sacks et al (2001) implies that a reduction to 65 mmol per day would provide even more benefit. A new soup formulation, of itself, will not achieve this. However, if similar re-formulation of other salty food products has similar taste outcomes, this goal will become easier to reach.
Australian Bureau of Statistics. 1999 National Nutrition Survey: Foods Eaten Australia 1995 ABS catalogue no. 4804.0 Canberra: ABS
Beard TC, Woodward DR, Ball PJ, Hornsby H, von Witt RJ, Dwyer T . 1997 The Hobart Salt Study 1995: few meet national sodium intake target Med. J. Aust. 166: 404–407
Bellisle F . 1998 Effects of monosodium glutamate on human food palatability Ann. NY Acad. Sci. 855: 438–441
Claiborne C . 1980 Craig Claiborne's Gourmet Diet p 45 New York: Times Books 1980
English R, Lewis J . 1991 Nutritional Values of Australian Foods Canberra: Australian Government Publishing Service
Favier J-C, Ireland-Ripert J, Toque C, Feinberg M . 1995 Répétoire Général des Aliments: Table de Composition 2nd Ed Paris: Institut National de la Recherche Agronomique
Holland B, Welch AA, Unwin ID, Buss DH, Paul AA, Southgate DAT . 1991 McCance and Widdowson's The Composition of Foods 5th ed. Cambridge: Ministry of Agriculture, Fisheries and Food and The Royal Society of Chemistry
James WP, Ralph A, Sanchez-Castillo CP . 1987 The dominance of salt in manufactured food in the sodium intake of affluent societies Lancet i: 426–429
Law M . 2000 Salt, blood pressure and cardiovascular diseases J. Cardiovasc. Risk 7: 5–8
Mattes RD, Donnelly D . 1991 Relative contributions of dietary sodium sources J. Am. Coll. Nutr. 10: 383–393
National Health and Medical Research Council. 1991 Recommended Dietary Intakes for Use in Australia Canberra: Australian Government Publishing Service
Okiyama A, Beauchamp GK . 1998 Taste dimensions of monosodium glutamate (MSG) in a food system: role of glutamate in young American subjects Physiol. Behav. 65: 177–181
Roininen K, Lahteenmaki L, Tuorila H . 1996 Effect of umami taste on pleasantness of low-salt soups during repeated testing Physiol. Behav. 60: 953–958
Sacks FM, Svetkey LP, Vollmer WM, Appel LJ, Bray GA, Harsha D, Obarzanek E, Conlin PR, Miller ER, Simons-Morton DG, Karanja N, Lin PH . 2001 Effects on blood pressure of reduced dietary sodium and the Dietary Approaches to Stop Hypertension (DASH) diet New Engl. J. Med. 344: 3–10
US Department of Agriculture, Agricultural Research Service. 1999 USDA Nutrient Database for Standard Reference, Release 13 Nutrient Data Laboratory Home Page http://www.nal.usda.gov/fnic/foodcomp (accessed 15 July, 2001)
Van der Heijden LJM, Hulshof KFAM, Langins JAE, van Oosten HM, Pruissen-Boskaljon JC, van Stratum P . 1996 Nevo Tabel: Nederlands Voedingsstoffenbestand Den Haag: Voorlichtingsbureau voor de Voeding
Yamaguchi S, Ninomiya K . 2000 Umami and food palatability J. Nutr 130: (Suppl 4): 921S–926S
Yamaguchi S, Takahashi C . 1984 Interactions of monosodium glutamate and sodium chloride on saltiness and palatability of a clear soup J. Food Sci. 49: 82–85
We are very grateful to Yolande Mou, Megan Cobcroft and Barbara Richli for helping us obtain European food composition data.
About this article
Critical Reviews in Food Science and Nutrition (2019)
Toxicology Mechanisms and Methods (2019)
Adoptable Interventions, Human Health, and Food Safety Considerations for Reducing Sodium Content of Processed Food Products
The Role of the Japanese Traditional Diet in Healthy and Sustainable Dietary Patterns around the World
Improved extraction methods for simultaneous recovery of umami compounds from six different mushrooms
Journal of Food Composition and Analysis (2017)