Umami Arising from Salt Rich Fermentations

  by Ben Reade.

Salt rich fermentations can be used to create high level of umami taste. Often the salt shuts off most possibility for microbial activity, so enzymes perform most of the protein breakdown required for umami taste. These enzymes may be naturally in the fermentation substrate (endogenous), or may be added (exogenous). To add, many salt rich fermentations, especially those which will later become sauces, regardless of how delicious they might be are brown in colour are not very photogenic – so photos today are really a bit random.

Salt rich fermentations are often used in creating food products that are both rich in acid (especially lactic acid, as many species of lactic acid bacteria manage to live in salt rich solutions) and in umami, due to the enzymatic activity on proteins (especially in legumes, meat and fish). Some yeast species are salt tolerant (eg. Zygosaccharomyces rouxii) meaning that some salt rich ferments are also quite rich in alcohols. In the following couple of posts there follows a brief look two of NFL’s more successful forays into salt rich fermentation, today, fish sauce and, definately worth returning for tomorrow (new post tomorrow morning), a Nordic variation on soy sauce. Happy umami hunting!

Fish sauce

One particularly rich source of umami tasting compounds is fish sauce.  NFL were inspired to experiment with creating several different fish sauces. Amano (1962) defines fish sauce as the “Clear brown liquid hydrolysate resulting form salted fish”. Although fish sauce is commonly associated with the cuisines of East Asia, there is a long tradition of fish sauce in Europe, especially in relation to the Roman ‘garum[1]. Analysis of fish sauces led Park et al (2002) to find that the most effective compounds for recreating the characteristic flavour of fish sauce were glutamic acid, followed by pyroglutamic acid and alanine.

Traditional fish sauces often take as long as one year to produce. At NFL a search was undertaken to develop a successful recipe for a sauce, using Nordic fish which not only tasted good but also did not take too long to make.

Enzyme rich materials can be used to encourage proteolysis, shortening the fermentation time (Chaveesuk et al, 1993. Aquerreta et al, 2002). For this reason, the fish digestive tracts, which are full of hydrolyzing enzymes, have been found useful in encouraging umami taste in fish sauce, as was common practice in garum (Grainger, 2010).

The first stage in this was the use of the fish’s digestive tracts, which, having been designed to break down the fish’s food (mostly other fish) and are also ideal for breaking down the fish in the fish sauce in autolysis. The most successful recipe trialed used fish digestive tracts and no fish flesh. This makes the recipe given below ideal for chefs wishing to use waste products (only guts of the freshest fish should be used).

It was also found that exogenous enzymes could be added using koji a widespread practice in Japan and elsewhere in Asia (Hariono et al, 2005). The most successful of recipes trialed at NFL included some amount of koji.

In later stages of fish sauce production, halophilic lactic acid bacteria (LAB) cultures produce lactic acid that adds to overall flavour and product palatability. Yongsawatdigul and Rodtong (2007) found that inoculation with various species of LAB led to higher production of free amino acids. At NFL, to encourage lactic acidification, the addition of acidified milk whey was included in the recipe. Addition of glucose may facilitate lactic acid production by lactic acid bacteria (Ijong and Ohta, 1999).

To encourage both LAB and enzymatic breakdown in the early stages of the fish sauce production, both of which have been shown to be slowed by excessive osmotic pressure,  salt was added in two batches. Half the salt was added at the beginning of the process, then the other half after 24 hours at 35°C. This was carried out to replicate a study by Aquerreta et al (2001), where it was found that best conditions were 10% Salt, 5% added at first, and 5% added after 24 hours at 35°C.

A double blind analysis of fish sauces made was carried out at Nordic Food Lab. Sauces were analysed for visual, aromatic, taste, flavour characteristics as well as aftertaste descriptors and hedonic rating. The following recipe came out as having the highest acceptability:

Fish Sauce Recipe

2730 g Mackerel (Scomber scombrus) digestive tracts

600 g Buckwheat koji (see FIG.

510 g Sea Salt

100 g Yogurt whey

First the above ingredients were weighed.

The salt was divided into two portions each of 255g.

All ingredients were combined leaving one portion of 255g of salt aside.

After 24 hours at 35°C, a second batch of 255g salt was added.

Resulting substrate was left at 35-40°C for 90 days.

Fish sauce was extracted by centrifuge and paper filtration.

The fish sauce recipe on the previous page is by no means definitive, but after experimenting with 8 different recipes, was the most successful. The aim now is to investigate further the use of koji as a source of exogenous enzymes and whey as a source of lactic acid bacteria for lactic acidification of the substrate. Do experiment with proportions and ingredients, Homer (the real one) wrote about how bad it smells, but how good it tasted. Think about all the valuable things Homer could have written about, if it smelled bad, it must have smelled REALLY bad, but the same also applies to it tasting REALLY good – so don’t be shocked it you need one of these (see photo below) what’s important to remember is your perception of a smell can change from horrible to delicious just with concentration. A fantastic example of this is the chemical Indole. Indole is at the same time the characteristic smell of orange blossom and, when at higher concentration the characteristic smell of human ‘solid waste’. This goes to show how with the same substance the perception can change completely with only a change in concentration – so fear not if it smells incredibly bad.

[1] For further information on garum the reader is referred to Sally Grainger (2010).


Amano, K. (1962) Influence of the nutritive value of fish with special reference to fermented fish products of southeast Asia, in Hee, N. and Kreuzer, R. (eds) Fish andNutritionFishing News, pp.180-200, London, UK.

Chaveesuk, R. et al. (1993) Production of fish sauce and acceleration of sauce fermentation using proteolytic enzymes, Journal of Aquatic Food Product Technology, 2 : 59.

Aquerreta, Y. et al (2002) Use of exogenous enzymes to elaborate the Roman fish sauce ‘garum’, Journal of the Science of Food and Agriculture 82 : 107.

Grainger, S. (2010) Roman fish sauce: an experiment in archeology, in Procedings of the Oxofrd Symposium on Food and Cookery ‘Cured, Fermented and Smoked Foods’ pp. 165-174, Prospect Books, UK

Hariono I, et al (2005) Use of Koji and protease in fish sauce fermentation, Singapore Journal of Primary Industries 32:19.

Yongsawatdigul, J. and Rodtong, S.N. (2007) Acceleration of Thai fish sauce fermentation using proteinases and bacterial starter cultures, Journal of Food Science 72 : M382.

Ijong, F.G. and Ohta, Y. (1996) Physiochemical and microbiological changes associated with bakasang processing – a traditional Indonesian fish sauce, Journal of theScience of Food and Agriculture 71 : 69.

Aquerreta, Y. et al (2002) Use of exogenous enzymes to elaborate the Roman fish sauce ‘garum’, Journal of the Science of Food and Agriculture 82 : 107.

About the author

My Name is Ben Reade, I’m a chef from Edinburgh, Scotland, and for the past 3.5 years I have been studying at The University of Gastronomic Sciences in Pollenzo, Italy. For my final thesis, I came to Nordic Food Lab to research many subjects where my varied interests inerlaced with those of the Lab. The research arose out of time spent at the Nordic Food Lab between 29 September and 22 December 2011. The aim is to describe NFL’s current research to both chefs and non-specialized readers, explaining and coding the creative and scientific methodologies employed during the research at NFL, exploring their application in food experimentation and innovation. Over the next month or so I will be breaking down this thesis into manageable blog-style chunks, this is chunk 6ish of around 25 I hope you find it interesting.