Why rennet anyhow?

[Lenomnom]

It seems any acid will coagulate milk, so why not just use lactic acid bacteria to run the pH down to the coagulation point? What is rennet doing that acidifying bacteria can't do?

[mikekchar]

Ha ha!  Don't get me started. 

Actually I'd write a big long detailed description of what's going on but I have a meeting in 3 minutes and then I'm out the door for the weekend.

Very quickly, making cheese with rennet is like making a house with bricks and mortar.  Making cheese with acid is like making a house with sand.  You can get the sand to stick together, but it will never have the texture of the bricks and mortar.  Essentially rennet makes it so that the protein can bind with dissolved calcium to form those bricks and mortar.  Acid just allows the proteins to collide and basically stick because they happen to be next to each other.

If you remind me (and nobody else does it) I can give you a more detailed explanation on Monday.

[awakephd]

Yep. What Mike said.

There are, of course, "lactic acid cheeses" which are set only by the action of the acid. The resulting curd is quite delicate, and must be drained a long time to reach the desired texture. Generally the final result is still quite soft. Note that yogurt is essentially a lactic-acid curd (thermophilic), so that will give a basis for comparison. (So-called Greek yogurt is simply yogurt that has been strained to let more whey escape, so it is closer to what you might be shooting for when making this type of cheese.)

To make anything resembling a hard cheese, you've got to give it some enzyme help. Rennet is a naturally occurring enzyme that helps young calves digest milk; if the milk did not curd up, it would go through them too quickly to be of benefit. One can achieve more or less the same effect with purely vegetable-based products, such as from thistles, but these do not contain quite the same mixture of enzymes and so will give a slightly different flavor profile.

[Lenomnom]

  So what you guys are saying, if I get this right, is that rennet induces stronger bonding between casein and/or casein fragments in the curd? With perhaps a more compact structure?

[mikekchar]

It's actually completely different chemistry.  It's one of the things that it commonly misunderstood (and there are many blog posts and even books that are quite misleading on this point).  I'm still not quite home from my mini-vacation, but here's a quick rundown.

Milk contains many different kinds of proteins, but as cheese makers we are interesting in a specific one called casein.  Casein is suspended in milk in large (microscopically, anyway -- you can't see it with the naked eye) balls called "micelles".  You can think of a casein micelle a bit like a big ball of bits of string -- where the role of the string is played by the casein protein.  On the outside of the ball is a slightly different protein called "kappa casein".  It covers the ball a bit like hair. The result is that you have a big hairy ball make up of tightly bound bits of string.

If you think of magnets, you may remember that magnets have a north and south pole.  If you get 2 bar magnets and put the north pole of 1 next to the south pole of the other, they will attract and stick together.  However, if you put the north pole next to the other north pole, then they will repell and you can't put them together.  The same goes for putting the south pole of one bar next to the south pole of the other bar.  (If you have fridge magnets you can try this by trying to stack them -- they will stick together one way, but the will repell the other way).  "Electric charges" work the same way (although we say "positively charged" and "negatively charged" instead of "north" and "south" -- but it's essentially the same thing.

I say all this because the hairy bits of the casein micelle (kappa casein) are positively charged and the big ball of string is negatively charged.  So the two stick together easily.  However, 2 casein balls can't stick together because both of them have this positively charged hair on the outside.  The result is that they float around in the milk without sticking together.  Incidentally, it is the casein micelles that make the milk white! The are distributed all through the milk because they are all repelling each other.

Let's first start with how rennet works, because it is actually simpler.  Rennet is an "enzyme".  Enzymes are chemicals that "cut" chains of other chemicals (usually proteins or carbohydrates).  Remember that we said that the casein micellle is a ball of tightly bound pieces of string, covered in hair.  Rennet gives the casein micelles a hair cut!  It's all it does.  All enzymes work best at certain temperatures and acidities.  For cheese making purposes, you can just remember that the warmer the milk and the more acid it is, the faster the rennet can give hair cuts (to a limit: too hot or too acid and the enzyme is destroyed, so be careful to follow cheese making recipes closely!)

What happens when we give the casein micelles a haircut?  Remember that the hair is positively charged and the ball is negatively charged.  We cut off the hair and we are left with the balls -- which are negatively charged.  Wait a minute!  Those can't stick together!  It's like nutting the south pole of one magnet next to the south pole of the other magnet!  What to do, what to do....?

Luckily chemistry comes to the rescue.  A "salt" is a chemical made up of 2 bits called "ions".  One of the ions is positively charged and the other is negatively charged.  They stick together nicely like magnets.  Normal table salt is sodium chlroide.  The positive "ion" is sodium (Na is the chemical symbol) and the negative "ion" is chloride (Cl2).  When you put it in water, something magical happens: it dissolves.  Water has a special property in that it's able to wedge most salts apart into their positive and negatve ions.  We call this "dissolving".  Dissolved table salt is just water with sodium ions and chloride ions floating around in it.

Milk has a special "salt" in it.  It's called calcium phosphate.  Normally there is a fair amount of it dissolved in milk.  Because it is "dissolved" it's found as calcium ions and phosphate ions floating around in the milk.   The calcium is positively charged and the phosphate is negatively charged.

Now to get back to the rennet: When we give the micelles a haircut, they become negatively charged.  Normally in milk, as I said, there is plenty of calcium ions floating around.  They are positively charged.  As soon as the micelle gets a haircut, the calcium ion sticks to it -- like a magnet.  The two bits are attracted to each other, so they will kind of "snap" together.  Then another casein micelle will "snap" to the attached calcium ion.  And then another calcium ion will snap to that micelled.  Over and over and over again, they just snap together.  Because the micelles are relatively big (microscopically speaking), there are gaps between the micells.  You can think of it like tinker toys (Google it if you aren't old enough to know what they are ;-) ) making a big cage.  It traps water and far inside the gaps.

And that's curds!

Now the acid formed curds.  You might be worried because I typed war and peace ^^^ to describe the "simple" case ;-)  Don't worry.  It's not that it's a long discussion, it's just that it's a bit harder to describe.    Remember that I said that the casein micelles float around (repelling each other) in the milk because of the hairy outside (the kappa casein).  It turns out that chemistry is weird (if you didn't already know).  When things get more acidic there comes a point where negatively charged things and positively charged things no longer repel/attract each other.  This is called the "isoelectric point" (literaly meaning "the point at which there becomes only one charge").

Just to make it clear, when we add enough acid to the milk, at some point the casein micelles no longer repel each other.  Neither do they attract each other, though.  If they happen to bump into each other, they will stick just because they happen to be sticky, physically.  At certain temperatures, they will get gummed up with the fat in the milk, etc and just happy clump together.  They aren't "bound" together, though.  If you push them (i.e. stir the milk) they will also happily move around -- like yogurt.

And, in fact, that is what yogurt is.  It's just the casein micelles (completely intact) sticking together.  If you've ever made yogurt, you may have seen the advice to "scald the milk" (bring it up to a temperature near boiling) before you make the yogurt.  The reason for this is that, as we know, casein is covered in hairy kappa casein.  This hairy outer covering mechanically makes it difficult for the casein to stick together.  If you heat up the milk, it "denatures" the protein.  You can think of it like scrambling an egg and not be too far off -- the protein is all there, but it is damaged and "scrambled".  Once you do that, you can never make a rennet cheese because the micelle is all messed up -- just like you can't make poached eggs after you scramble them ;-)  That's why you can't make rennet cheese with UHT (Ultra High Temperature pasteurisation method) milk.

If we scald the milk, we can damage the micelle, but it will still pile up nicely when we add acid to the milk (and hit the isoelectric point).  In fact because the hairy kappa casein is basically destroyed, the damaged micelles can sit closer together and so the resultant yogurt is firm.  In fact, it looks *very* much like rennet made curd.  You can even cut the curd and whey will drain out.  But if you do *anything* with it, the piled up, damaged casein, will just slide around and collapse like a sand castle collapsing into a pile of sand if you poke it too much.  There is no way to make a solid cheese like this (and believe me I've tried!)

So yogurt will never make a solid cheese because the casein micelles aren't bound together, right?  Well, that's not completely true.  There is a trick.  If you add acid slowly to milk, it sets up into a fluid yogurt with only a few casein micelles sticking to each other.  However, if you add acid quickly, the curd will coagulate into something quite hard.  You can actually press this and make a hard cheese (I have done it *many* times).   You can *not* "cook" this curd because it is still very delicate -- all the fat will come out if you try.  We don't have our nice calcium built cage to hold in that fat.  Also if you try to press the curd very hard, you will similarly loose all the fat.  But there is another trick.  If you hold the cheese above the melting point of the curd, it *will* melt and stick together.  If you then quickly move the melted curd into a mould, it will set up under it's own weight (or with a *tiny* amount of pressure).  You can age this cheese (however it will never form a rind, so it's a bit tricky to do so).

Why does adding acid quickly change whether or not the curd is soft or hard?  I don't know.  Since the message is already too long and I have to do some work, I'll leave that teaser.  I've got some idea what might be happening and am happy to speculate, but I can discuss it later with anyone who is interested.

I don't have time to edit this post, so forgive my errors!

[Lenomnom]

It's actually completely different chemistry.  It's one of the things that it commonly misunderstood (and there are many blog posts and even books that are quite misleading on this point).  I'm still not quite home from my mini-vacation, but here's a quick rundown....

Well this is brilliant this is. I now have a working model of cheese on the molecular scale to play with. It even leads me to a novel prediction; Milk should set a weak yoghurty curd if treated with massive quantities of salt and gentle heat. Don't know if anyone would want to eat it though!

A couple of questions linger, as they always do.

First, is there a structural difference between kappa casein and the other type? I've heard there are genetic variations in the kappa caseins and it sounds like there are two variations of kappa casein and they follow Mendelian inheritance principles. But do kappa caseins as a class have some broad difference from the caseins in the center of the micelles? It sounds to me like they are behaving like soaps, so I expect they are quite polar. The central caseins seem to be behaving like some type of oil, i.e. not polar at all.

Second, is it the case that this isoelectronic point is just the point at which there are so many free ions in the solution that they can cluster near the micelles and effectively cancel the charge of the kappa proteins?

Third, in an acid set cheese, is the stickiness you speak of just hydrogen bonding?

Fourth, where's the fat? I thought those micelles were primarily made of fat, not protein. Are the central caseins dissolved in milk fat globules?

btw I have heard from one lady and her cow (who is named Ladybug and to whom the internet of cheese owes a great debt) that one of the kappa casein variants is much better for cheesemaking and curd setting than the other. Since Ladybug is homozygous for the good type, her owner is one happy cheesemaker.

ps the only significant typo I saw was the use of the word "far" where "fat" was intended.

pps Thanks again, brilliantly clear post, worthy of a sticky if you ask me.

[mikekchar]

Good questions.  I'll try to answer, but I'm still learning about this as I go.

Wikipedia page for kappa casein: https://en.wikipedia.org/wiki/K-Casein  Wikipedia page for casein: https://en.wikipedia.org/wiki/Casein  Questions about the structure of micelles is hard to answer because it's still a topic of active research.  In other words, nobody really knows exactly how it is set up.  In fact if you try to find out exactly why cheese melts or stretches, you will be on a nice ride of scholarly papers on the subject

Casein is largely hydrophobic (it repels water).  I'm not at all knowledgeable about soaps (especially how they work at a molecular level).  However, I don't think it is generally related to the way soap works.  I'll be honest and say that I don't really understand how anything is hydrophobic.  Just to give a bit of background, though, water is a "polar" molecule. It has one oxygen atom and 2 hydrogen atoms and is shaped a bit like a boomerang with the oxygen in the middle.  The hydrogen is positively charged and the oxygen is negatively charged.  The way water "dissolves" salt is that the negative ions are attracted to the hydrogen and the positive ions are attracted to the oxygen.  This is enough to wrench the salt apart and for the ions to wander away. 

Kappa casein (k-casein) is not polar as far as I know.  It is positively charged.  "Polar" means that there is a different charge on one end than the other.  The micelle itself is round (and very, very big compared to other molecules -- enormous, really) and the whole thing will be positively charged when the K-casein is attached.  The whole thing is negatively charged when the k-casein is removed (cleaved).  I suspect its size and charge have something to do with the reason it is hydrophobic, but I don't really know.

k-casein is not particularly interesting with respect to cheese making because we cleave it off and it gets drained away in the whey.  I wouldn't get too worked up in the name, either.  Proteins are just large chains of molecules with nitrogen in them.  Proteins are made up of building blocks called "amino acids".  The way these amino acids are put together in the protein determines what kind of protein it is.  K-casein is just a casein molecule with a slightly different "shape" (combination of amino acids).  The rest of the micelle contains some other shapes as well (and since they got to kappa in the Greek alphabet I guess there are alpha, beta, gamma, delta, epsilon, zeta, eta, theta and iota caseins at the very least .

Because it is a marketing term, some people know about A1 versus A2 beta casein.  I have absolutely no idea what role beta casein plays (if anything) in cheese, but some people are allergic to A1 beta casein.  A2 is just the same as A1, but has a different amino acid somewhere in the chain (Wikipedia has more info in the casein article).  I suspect this is what your farmer friend is talking about.  Maybe not???  Anyway, Jersey cows have more A2 beta casein while most other cows have mostly A1 beta casein. 

I'm not entirely sure how the isoelectric point works.  Here's another Wikipedia page: https://en.wikipedia.org/wiki/Isoelectric_point  My very loose understand is this: "acid" has a very specific definition.  I don't want to write a book on ionic chemistry, but essentially water will "dissociate".  In other words instead of H2O, it will split into an H+ ion and an OH- ion.  In water that has a pH of 7 (neutral), these ions are in balance.  When they get unbalanced the water becomes either acidic or basic.  More H+ ions than OH- ions and it is acid.  More OH- ions than H+ ions and it is basic.  This is literally what the pH scale measures.  It's important to understand that the H+ and OH- ions will react with other substances in the water and then "precipitate out" (i.e. come out of solution).  This strips either H+ or OH- ions from the the water.  Normally if we add an acid to water, that acid reacts with the OH- ions in the water, creating a salt.  This leaves more H+ ions than OH- ions, making the water more acidic.

Some judicious cutting and pasting from Wikipedia: "The pI value can affect the solubility of a molecule at a given pH.... Amino acids that make up proteins may be positive, negative, neutral, or polar in nature, and together give a protein its overall charge. At a pH below their pI, proteins carry a net positive charge; above their pI they carry a net negative charge."

In all of these ionic systems there is a free exchange of ions.  Things bind a positive ion and then dissociate (and the same with the negative ions).  Everything is in flux and it can be a bit hard to say "What is it now".  These big proteins just keep exchanging ions and the net result is either having an excess of positive charges (in which case the proteins are positively charged) or having an excess of negative charges (in which case the proteins are negatively charged).  But in any case, when the  pH hits the pl for the protein, it no longer has a excess of either positive or negative charges.  It's neutral.

Just a quick sentence to say that I've often wondered how this affects whether or not the protein is hydrophobic.  Not sure... because I don't know if that is related to charge at all.

Sorry for the crap explanation.  I don't understand it myself...

You asked what about the fat?  Fat is *not* held in the micelle.  The micelle is just a tightly bound ball of protein and calcium phosphate salt (which may or may not play a role in keeping the micelle together depending on what paper you read).  The fat exists as globules floating around in the liquid.  When the casein micelles stick together (either mechanically or through boding with calcium) there are spaces in between.  That's where the fat gets trapped.  This should give you a good idea of how big these micelles are: they are big enough that when you stack them all up, the space between them is enough to trap globs of fat!

Finally about the stickiness: no idea.  I don't believe it is hydrogen bonding, but then I don't know why *anything* is sticky   I actually suspect that nobody has studied the reason why yogurt gels as it does, given the fact that nobody really knows how a casein micelle is put together.  It might sound weird, but I remember about 30 years ago going to a physics seminar discussing the topic of the formation of the head on a beer.  Before that nobody knew how it worked in detail.  The same is true about a lot of the physics and chemistry of cheese making.  It's difficult stuff to study, I think.

Hope that helps.  I even hope more that someone will come by and fix my mistakes :-)

[Lenomnom]

 Ok I did my reading and a few more things are becoming clear to me. I had a fair understanding of how micelles form in soap - oil - water systems and thought that the casein micelles might work the same way, but it's only partially true.

 One thing I didn't mention was that it occurred to me you could probably make some kind of plastic from casein, and I was delighted to see it has been done.

It's pretty amazing that we still don't have much of a grip on casein micelle structure. It looks like the only sure way to determine if a cheese is meltable is to make it and try to melt it. Our best theories provide only an unreliable guide.

I can maybe explain hydrophobicity a bit. Water is a polar molecule, the oxygen hogs the electrons a bit meaning that the molecule is a bit negative on the oxygen end and a bit positive on the hydrogen ends. So we have opposite charges which attract. In liquid water this causes the molecules to form loosely linked, ever breaking and reforming networks.  Simply put, water molecules stick to each other quite a bit. Now suppose we introduce a molecule into the water that isn't like water, but instead has nice uniform charge distribution with no positive or negative hot spots. Most oils qualify. This new molecule can't play the sticky game.

   Thought experiment: what would happen if you had a hockey arena with 1000 people in it, 500 of whom were chronic huggers and the other 500 are hug refusers? The huggers would all stick together in a lump and the non huggers would be excluded. Like oil and water, like polar and non-polar, they tend to separate. Any polar molecule can play the sticky-huggy game, but no nonpolar molecule can play. Interestingly, there are some molecules that have a charge hotspot at one end but none at the other. These hybrids are called soaps.

(btw I'm just doing a running commentary as I scroll through your previous post here)

   I have to disagree with you about caseins and polarity. According to the wiki caseins occur in milk as calcium salts, calcium ions have a +2 charge and therefore caseins must be negatively charged. It seems likely to me that that charge is localized around some amino acids and not others, which would lead to an uneven charge distribution - that's polarization. Which raises the question, why doesn't casein just dissolve in water? I would guess it is because the molecules are large, the charges are scattered about the molecule, so overall the casein molecule plays the sticky-huggy game badly. It might be the opposite, maybe they play the game really well, but so long as there is a significant difference, one will agglomerate and exclude the other. Maybe.

  Thanks for pointing out the difference between fat globules and casein micelles. I note that if casein is quite polar, it would explain why it isn't all found in the fat globules. Fats are highly non polar. Your description of the casein micelle is pretty good, but wiki points out that there are significant amounts of water in the casein core. How odd. Or is it?

     Please fire back any objections you have, I'm still forming a picture and any critique is helpful.

   I have to agree with your closing comments. The micro world of casein and curds is indeed complex, sometimes baffling with many conflicting physical and chemical processes at work. It's a battle royal down there and it's a wonder we get cheese out of it almost every time. If you had just told me the chemistry and nothing else, I would have expected the result to typically be something like set plaster. Chalk, not cheese. Or a polymerized plastic. It's a wonder the stuff is edible, let alone tasty!

[mikekchar]

Interesting.  Casein definitely *is* negatively charged as you say.  It's the k-casein that is positively charged.  I haven't looked into why that is.   I'm really at the edge of my understanding and can't really contribute anything more without more study.  It's discussions like this that make me learn stuff.  Thank you! :-)

[awakephd]

Well, the chemistry is over my head ... but nonetheless I'm getting a *charge* out of the discussion!

[Lenomnom]

Well, the chemistry is over my head ... but nonetheless I'm getting a *charge* out of the discussion!

"Are you sure?"
"Yes, I'm positive."

 

[Susan38]

Wow, this discussion is a great example of why I joined this forum!  I had no idea that the cheesemaking process at the molecular level is still somewhat poorly understood/controversial.  Thanks for explaining what is known about it in a way that I was able to follow (for the most part).  Chemistry and physics aside, could there be some biological interactions that are playing roles (other than producing acid, flavors, and gases) that have yet to be recognized?  I understand that with vat pasteurization, there are still some bacteria (and ??) in the milk, along with the culture bacteria that are added in some cheese recipes.  Whatever all is going on in the formation of these cheeses, it does seem to involve quite a complex process and maybe one that will never be fully understood.  And maybe that is part of the fun and intrigue with cheesemaking?

[mikekchar]

Yep.  I totally agree.  I've been finding that even more than my brewing hobby (which I sadly had to give up because it is illegal in Japan) making cheese makes me feel like a mad scientist    At the same time there is a kind of magic that happens where you can understand the processes as much as you like, but then when you taste the cheese it's really hard to believe that something so delicious can be the result of what you did.

With respect to the bacteria, its main role is in supplying the acid for the equation.  Milk is full of a sugar called lactose.  The bacteria eat the lactose and produce an acid called lactic acid.  Lactic acid has the formula: CH3CH(OH)CO2H and when it dissolves it produces the lactate ion which is exactly the same thing with the last H missing :-)  That H is what makes the milk acidic.

However, if that was all it was, cheese would be boring.  Instead the bacteria produce other chemicals as well which produce a variety of flavours.  There are lots and lots and lots of lactic acid producing bacteria and they all have slightly different characteristics.  There are also lots and lots and lots of other bacteria that can live in milk with lactic acid bacteria and these produce lots of different flavours as well.  Not only that, but when the bacteria dies (which it will eventually because it runs out of food), there are enzymes in the cell walls.  Remember that enzymes cut apart proteins and carbohydrates.  These enzymes break down the proteins in the cheese leaving smaller amino acid chains.  Each size/shape of amino acid chain has  a different flavour!  So it's just the cornicopia of flavour producing effects.

For me, this is the exciting part of cheese making.  You use science to guide your understanding of what's going on, but you use art to work with the resulting palatte you get.  Every day and every cheese is a learning process.  It can't get much better than that :-)

[Lenomnom]

 Yup cheesemaking is still as much an art as it is a science, or maybe a strange biological version of alchemy.When you consider that just changing what the cow eats can have a major effect on the outcome... and the time of year and the selection of plants available to the cow matter as well... to say nothing of the bacteria on the exterior of the cow, which we strain to eliminate but never quite succeed at. The only option is to take an empirical approach. Sure you can let theory guide you but there are no guarantees at all.


   One of my favorite puzzles is slip skin in Brie and Camembert. Even Gavin Webber has not completely solved it, last I heard. I think it has something to do with the strain of mold and the things that affect mold growth, such as temperature and humidity and salinity and pH. But I could be totally wrong and it might one day turn out to be due to insufficient infiltration of carbon dioxide from the atmosphere or something else no one is paying attention to today. Any bets on an unknown symbiotic mold?
 
   So which will mankind master first, fusion power, warp drive or a good consistent Brie anyone can make?

   

[Lenomnom]

Interesting.  Casein definitely *is* negatively charged as you say.  It's the k-casein that is positively charged.  I haven't looked into why that is.   I'm really at the edge of my understanding and can't really contribute anything more without more study.  It's discussions like this that make me learn stuff.  Thank you! :-)

I found this in an abstract to a technical paper on k-casein:

Biochimica et Biophysica Acta [01 Mar 1977, 491(1):93-103]
Type: Journal Article
DOI: 10.1016/0005-2795(77)90044-7
Abstract
1. A description is given of the fractionation of kappa-casein on DEAE-cellulose using a pH gradient. With this method an improved separation of the kappa-casein components with a higher negative charge is obtained. 2. It is shown that at least one of the kappa-casein fractions has a second phosphate ester group. The heterogeneity of kappa-casein therefore is not exclusively caused by a varying N-acetylneuraminic acid content.

This says to me that the kappa casein is also negatively charged. Whaddaya think, Mike? Are you up for a reference fight?


edit: eep i forgot to put in the source  url: https://europepmc.org/abstract/med/849471