A while ago a reader sent me a very interesting question regarding a gelled seafood sauce. It is made by mixing tomato ketchup with horseradish and his question was very simple: Why and how does this sauce gel? He speculated about pectin (which is present in tomatoes), but wondered why ketchup then doesn’t gel on it’s own? And he also noted that horseradish ground with water does not have any gel like properties. So how come they can form a gel when mixed together?


Grated horseradish

The first thing that came to my mind was a previous blogpost on tomato gels with the pectin that’s there. Pectin in tomatoes is highly methylated (HM), meaning that a lot of sugar would be required for it to gel and that gelling is not promoted by calcium. But if it is mixed with juice from carrots or oranges which contain the enzyme pectin methylesterase (PME), the methyl groups are cleaved off (as shown below) to yield a low methylated (LM) type of pectin which will gel more easily, especially in the presence of calcium ions. Could something similar be the case in the gelled seafood sauce?

Once horseradish is cut, enzymes start to break down sinigrin to release allyl isothiocyanate (mustard oil) which is responsible for the pungent taste and the irritating effect on the eyes and sinuses. In biochemistry, horseradish is best known for an enzyme called horseradish peroxidase. I’m not sure if this is the enzyme that is responsible for the degradation of sinigrin, but adding together the bits and pieces my best guess is that some enzyme in horseradish does more or less the same thing as pectin methylesterase, cleaving of methyl groups to make the pectin more prone to gel. I haven’t been able to find any papers on this though, so if any readers know more about his – please feel free to fill me inn! And can you think of other foods where horseradish advantageously could be used for gelling?

Using an Aeropress as a pressure filter to obtain a horseradish extract

Before writing this blog post I wanted to test the gelling, so I took a pieces of horseradish, peeled it and grated it. The gratings were quite dry so I decided to mixed them with some water and then filter the mixture to obtain a horseradish extract. The Areopress coffee maker turned out to be perfect for this (as shown in the picture above). I then mixed the extract with approximately 6-8 times the amount of ketchup and left it in the fridge to gel. A before-and-after picture of the ketchup mixed with the horseradish extract is shown below. If I would make this again however, I’d probably not bother about filtration – instead I would use a food processor with knives to break up the cells in the horseradish as much as possible to maximize the release of the intracellular enzymes.

Ketchup and horseradish extract immediately after mixing (left) and after a night in the fridge (right)

The suggested recipe I received with the question was as follows (more can be found by googling “seafood sauce” or “cocktail sauce” in combination with ketchup and horseradish):

Gelled seafood sauce
250 mL horseradish
4 L ketchup to
25 mL lemon juice

Grate/grind horseradish with a little water. Mix with ketchup. Adjust with lemonjuice (and possibly salt) to taste. Refridgerate. The gelling doesn’t happen until a day or so later.

This month’s round of TGRWT is hosted by Pablo over at Medellitin, and the foods to pair this time are tomato and black tea. As always you can find instructions on how to participate in the announcement post. If you are new to TGRWT (which stands for They Go Really Well Together), check out the round-ups of the previous 18 rounds! And if you are chemically inclined, you may want to read on to learn more about the compounds behind this months pairing.

With a little help from Douglas Baldwin (whom I interviewed about sous vide recently) I’ve been able to pinpoint the compounds which occur naturally in both tomato and black tea, according to The Good Scents Company website:

(E)-2-hexen-1-al, (E)-2-hexen-1-ol, (E)-2-hexen-1-yl acetate, (E)-2-nonen-1-al, (E)-geranyl acetone, (Z)-2-hexen-1-ol, (Z)-3-hexen-1-al, (Z)-3-hexen-1-ol, (Z)-3-hexen-1-yl acetate, 1-octen-3-ol, 1-penten-3-ol, 2,4-decadien-1-al, 2-hexen-1-ol, 2-methyl furan, 5-methyl furfural, ammonia, butyl alcohol, butyraldehyde, butyric acid, citronellol, dihydroactinidolide, dimethyl sulfoxide, dimethyl trisulfide, ethyl hexanoate, gamma-hexalactone, gamma-valerolactone, geranic acid, hexanal, hydrogen sulfide, isoamyl alcohol, isovaleraldehyde, isovaleric acid, linalool oxide, methyl ethyl ketone, ortho-guaiacol, propionaldehyde, valeraldehyde

Now this might seem impressive, but as I’ve touched upon previously it is highly uncertain that all of these compounds actually contribute to the flavors of tomato and black tea. Many are probably present at concentrations well below the individual odor thresholds. To alleviate this one preferably needs odor activity values. The closest I came for tomatoes was the mention (free pdf) of a “model” tomato paste with the following compounds:

(E)-beta-damascenone, 2-phenylethanol, 3-methylbutanal, 3-methylbutyric acid, 4-hydroxy-2,5-dimethyl-3(2H)-furanone, 4-hydroxy-4,5-dimethyl-2(5H)-furanone, 4-vinylguaiacol, 5-ethyl-4-hydroxy-2-methyl-3(2H)-furanone, acetic acid, butyric acid, dimethyl sulphide, eugenol, linalool, methional, methylpropanal, vanillin

And for tea (both black and green) there is a complete PhD thesis available for download (in German). The following compounds in black teas had high FD (flavor dilution) values:

(E)-2-nonenal, (E)-beta-damascenone, (E,E)-2,4-decadienal, (E,E,Z)-2,4,6-nonatrienal, 2-phenylethanol, 3-hydroxy-4,5-dimethyl-2(5H)-furanone, 3-methyl-2,4-nonandion, 4-hydroxy-2,5-dimethyl-3(2H)-furanone, beta-ionone, geraniol, linalool, phenylacetaldehyde, phenylacetic acid, vanillin

Comparing the two latter lists, we get the following shortlist for odorants present in tomato (paste) and black tea which contribute significantly to their aromas:
(E)-beta-damascenone, 2-phenylethanol, 4-hydroxy-2,5-dimethyl-3(2H)-furanone, linalool, vanillin (shown below). The fact that none of these are included in the data from The Good Scents Database illustrates my point about using OAVs to evaluate flavor pairing.

References:
Werner Grosch “Evaluation of the Key Odorants of Foods by Dilution Experiments, Aroma Models and Omission” Chem. Sens. 2001, 531.
Schuh, Christian “Wichtige Aromastoffe in schwarzem und grünem Tee (Camellia sinensis)”, PhD disseratation, TU München, 2004.

A recent article (found via Harold McGee’s News for curious cooks) featuring Heston Blumenthal as a co-author emphasizes the huge difference in glutamic acid contents between the flesh and pulp of tomatoes. Glutamic acid and it’s sodium salt (mono sodium glutamate or MSG) are responsible for the characteristic umami taste. On average the flesh contains 1.26 g/kg glutamic acid whereas the pulp on average contains 4.56 g/kg glutamic acid. Similar differences are found for several nucleotides which posess similar taste qualities. These differences can explain the perceived difference in umami taste between the flesh and pulp of tomatoes – and is worthwhile considering when cooking.

Those concerned about food with added MSG should read the chapter about MSG in John Emsley’s excellent book “Was it something you ate?”. First thing to note is that you can’t be allergic to MSG because our body needs glutamic acid to function properly. Emsley retraces the history of the Chinese restaurant syndrome (CRS) back to it’s roots in 1968 when a letter was published (R.H.M. Kwok, New Engl. J. Med. 1968, 278, 796) describing a series of symptoms experienced after having eaten at a Chinese restaurant. To make a long story short, in 1993 Tarasoff and Kelly reviewed previous studies and conducted a double blind test which led to the following conclusion:

… ‘Chinese Restaurant Syndrome’ is an anecdote applied to a variety of postprandial illnesses; rigorous and realistic scientific evidence linking the syndrome to MSG could not be found.

Following the publication, a critical reply was published by Adrianne Samuels, to which the authors have replied.

Anyway, it was in John Emsley’s book that I first read about the record levels of glutamic acid found in parmesan cheese: 12 g/kg! That’s nearly three times the amount found in tomato pulp. In some cheeses there is so much that it crystallises out in small white crystals visible to the naked eye. Think about this when you sprinkle your food with parmesan. And if you ever wondered why Italian food tastes so nice, now you know that MSG is one reason (but of course not the only one …).