By Eben van Tonder, 22 June 2026
Executive summary
Key findings.
- Six years ago, I made reformed ribs using imported riblets, adding meat back on with transglutaminase.
- The first product failed at the join between the reformed meat and the rack, because there was a lot of exposed fat on the bone and not enough meat.
Recommendations.
- The better procedure would have been to use the Klebemasse technique, with proper protein extraction and no connective tissue or fat in what was added to the ribs. In other words, use primals and not trim, with a rind emulsion (HeatCut Hautstoss) for bite and yield, the enzyme only for green strength, and a gentle cook to set the whole structure.
- Basic process. Thaw and cure the meat. Make the HeatCut Hautstoss. Make the Klebemasse. Mix the reformed meat layer. Form the rack from the sheets and add the Klebemasse and the meat. Press and set cold. Steam cook. Glaze and pack.
- Final expected yield. The added part on top is most of the rib, about 60 percent, and it is made with 40 percent water added on its own lean, most of that water held in the Hautstoss as gel. On the worked example rack the final glazed and packed rib weighs about 2420 to 2530 g from about 1600 g of lean meat, a net gain of about plus 50 percent over the lean, or about plus 25 percent once the bone is counted. The glaze at 10 percent roughly offsets the cook loss.
The phone call
Six years ago I made a reformed rib at Van Wyngaardt’s in Johannesburg. I took imported riblets, thin sheets of rib meat, joined them, and bound a layer of seasoned pork on top with a transglutaminase from the Ligar range. On the bench it looked good. It held, it sliced, and it photographed well, so I was pleased with myself.
One day the client called me. One of his clients bought the product, taken it home, and found the binding did not hold. The meat on top had come away from the rack beneath it. After only a short while working with master butchers in Austria and Germany I finally understand why.
The faults were simple. The riblets were trim, so they had a lot of exposed fat and not enough meat, the bind was weak, and the colour was uneven. So here is the complete method I should have used, cured, built on a proper rack, cooked, glazed and packed, step by step.
Where the first rib failed
The first rib failed at the join between the reformed meat and the rack. The riblets were imported trimmed ribs, so they had a lot of exposed fat on the bone and not enough meat, therefore the join had almost no clean protein to bind. Microbial transglutaminase, first isolated by Ando and colleagues[1], builds a covalent crosslink between glutamine and lysine residues on proteins, the bond set out in the transglutaminase chemistry of Folk[2]. It works on protein, however it does nothing on fat. Kurth and Rogers showed the enzyme crosslinking muscle protein to other proteins, never to fat[3]. The salt soluble protein that gives the tack and feeds the enzyme has to be extracted from lean meat, which the work of Hamm[8] and of Offer and Knight[9] explains, and a fatty trim face does not provide it. There was probably enough enzyme. More would not have helped, because the enzyme had nothing clean to hold.
The better procedure is the Klebemasse technique, with proper protein extraction and no connective tissue or fat in what is added to the ribs. In other words, use primals and not trim, and if any trim is used keep it close to zero. The lean is cured first, so the colour, the bind and the safety are set, a Klebemasse brings extracted protein to the join, a rind emulsion adds bite and yield, and the enzyme does only the green strength before a gentle cook sets the whole structure.
The material you are building from
The sheets you build from are mostly muscle, with fat only as seams, the white lines running through the meat, not a fat layer sitting over the bone. So trimming the surface fat and scoring lightly reaches muscle across most of the face, because the bulk of each piece is lean. That is far better material than the fatty trim that failed before.
The bond does not need every square millimetre either. The lean only has to be the clear majority of the face, with the bonded area spread out, so that no single fat seam is left carrying the load. Scattered fat seams are tolerable. The first version did not fail because a few patches were fat, it failed because there was too little meat in total, so the bonded fraction was simply too small.
The shape of the new process
Thaw and cure the meat. Make the HeatCut Hautstoss. Make the Klebemasse. Mix the reformed meat layer. Form the rack from the sheets and add the Klebemasse and the meat. Press and set cold. Steam cook. Glaze and pack. Each step is below.
Step 1. Thaw and cure
The riblet sheets arrive frozen, so they have to be thawed first, and thawed correctly. Thaw them under refrigeration rather than at room temperature, so they stay cold and safe, to a workable point of about minus 2 °C to plus 2 °C, firm but no longer frozen. A bond will not form on a frozen face, therefore this step decides much of the result.
Curing does three jobs here. The salt draws out the myofibrillar protein that the bond and the cook both need, so the cure primes the meat for binding. The nitrite develops the cured pigment, which sets on cooking to the stable pink of nitrosylhemochrome, so the colour stays even through the cook instead of the grey that looked poor on the first product. The nitrite also controls Clostridium botulinum and protects the fat. Honikel sets out these functions of nitrite[11].
Cure the thawed sheets, and cure the primal lean that becomes the reformed layer and the Klebemasse. Add Origins Cure at 25 g per kg of meat. It is formulated to a compliant ingoing nitrite level, and it also has the phosphate and the cure accelerator in it, so it does the whole cure in one addition.
Method. Rub or mix Origins Cure into the meat at 25 g per kg. Apply it to the sheets as they thaw, so they cure and develop a tacky, salted surface by the time they are ready. Hold everything under chill at 0 to 4 °C. The thin sheets cure quickly, however an overnight hold does no harm, and the primal lean should rest overnight so the cure reacts and equilibrates. That rest also lets the nitrite and the accelerator react before the enzyme arrives, which protects the enzyme, because its active site cysteine is sensitive to these reactive species, and because salt, phosphate and temperature all bear on how well the enzyme then binds[7]. The free water that does not go into the Hautstoss, about 20 percent of the added water, goes in later when the layer is mixed, held by this cured lean and the carrageenan. Blot any surface drip from the thawed sheets before you build, because a water film at the join weakens the bond.
Step 2. Make the HeatCut Hautstoss, the rind emulsion
The HeatCut Hautstoss is a rind emulsion. In other words it is pork rind, cooked soft and then chopped fine with water into a smooth paste. Its collagen gelatinises in the cook and sets to a gel on cooling, so it holds water as a gel, adds bite, and helps the added mass stay together. This is where the collagen in the build belongs, put in on purpose, rather than as connective tissue left in the meat. A rind emulsion is made hot, unlike a cold meat batter, because the collagen has to be soft and hydrated before it will emulsify.
The ratio decides how much water the gel will hold. One part rind to two parts water, about 67 percent water, is the working optimum, because it sets firm enough to handle and slice yet still holds a great deal of water. A firmer gel at one to one and a half holds less water. A wetter gel beyond one to two and a half turns soft and starts to weep. So one to two is the balance.
| Ingredient | Amount |
|---|---|
| Pork rind, cleaned of hair and excess back fat, raw | 1000 g |
| Water, for emulsifying, hot | 2000 g |
| Origins Hautstoss Mix | 20 g per kg of finished emulsion |
Method, step by step.
- Scald and clean the rind, then remove the loose subcutaneous fat, because fat does not belong in a collagen gel.
- Simmer the rind in water until it is fully soft and pinches apart easily, roughly sixty to ninety minutes at a gentle simmer near 90 to 95 °C, or about thirty minutes under pressure. The aim is gelatinised, tender collagen.
- Drain, keeping some of the hot cooking liquid.
- While the rind is still hot, chop it in the bowl cutter with the 2000 g of hot water, using some of the reserved cooking liquid, until it forms a smooth and viscous emulsion. Keep it hot through this step, in the region of 60 to 80 °C.
- Add the Origins Hautstoss Mix near the end.
- The emulsion sets to a gel as it cools. Chill it, or freeze it in slabs for storage as a HeatCut base, and soften it again gently before use.
This batch sets to a firm gel at about 67 percent water, so a small weight of it holds a large weight of water in the final mix.
Step 3. Make the Klebemasse, the binding paste
The Klebemasse is a binding paste, and the German name means glue mass. In other words it is lean meat chopped cold with a little water, the salt already in it from the cure, worked until the myofibrillar protein comes out and the mass turns sticky. That tack is protein glue. It brings already extracted, sticky protein straight to every join, and it gives the enzyme a protein rich layer to crosslink, which is what the first rib lacked. Kuraishi and colleagues showed that salt with the microbial enzyme binds raw meat pieces cold, at chill temperature in about two hours, which is the behaviour the Klebemasse relies on[5]. It goes into the seams between the sheets and across the whole top face, so make it to the ratio below and scale the batch to the area of the rack, spreading it as a thin film of about two millimetres.
| Ingredient | Amount |
|---|---|
| Cured primal lean pork, from the cured batch, well chilled | 10 parts |
| Iced water | 1 part |
| Microbial transglutaminase (Ligar) | about 1.0 percent of the paste, set to the declared activity |
As a starting batch that is 200 g of cured lean to 20 g of iced water, scaled up to the area the rack needs.
Method. Chop the cured lean cold with the iced water until it turns into a bright, sticky paste. The salt it needs is already in it from the cure, so add none. Keep it cold so the fat does not smear. Mix the transglutaminase in just before you spread the paste, because once it is hydrated it starts to crosslink, so make only what you will use straight away.
Step 4. Mix the reformed meat layer
This is the meat that goes on top of the rack, and it must be primal lean, not trim, because the bind needs clean protein with no fat and no connective tissue in the meat itself. The collagen in this build comes on purpose from the Hautstoss. It is made with 40 percent water added on this lean, and the split of that water is more important than the amount. Put about 80 percent of it into the Hautstoss as gel bound water, and leave only about 20 percent as free water in the cured meat. Gel bound water survives the cook far better than free water, because the muscle lattice shrinks and pushes out free water as it heats, while the collagen gel holds its water as a set gel[9][10]. Pushing toward 90 percent in the Hautstoss is possible, however it needs either more rind, which turns the bite gummy, or a wetter gel, which weeps, so 80 percent is the balance. Prove the level with a cook out test. If you must use any trim, keep it close to zero and cold, because fat smears, coats the protein, and blocks the bind.
| Ingredient | Amount |
|---|---|
| Cured primal lean pork, minced or diced, chilled | 1000 g |
| HeatCut Hautstoss, softened | 480 g, about 320 g of it bound water, the 80 percent |
| Iced water | 80 g, free water, the 20 percent |
| Origins Bind System | per the pack, to lock the free water |
| Origins Spice Mix | per the pack, for the rib character |
| Microbial transglutaminase (Ligar) | about 0.5 percent of the layer, for green strength, set to the declared activity |
Method. Mix the cured primal lean first, working it until the surface turns tacky, because that tack is the salt soluble protein doing its work. Blend in the Origins Bind System, then the softened Hautstoss and the small free water, then the Origins Spice Mix. Add no further salt, because the cure has it already. Keep the mass cold so no fat smears. If a little fat is wanted for mouthfeel, add a small amount of cold back fat away from the joins, however keep it small, because the bind wants clean lean. Add the transglutaminase last, just before you build, because once it is hydrated it starts to crosslink. It gives the big added mass its green strength so it holds shape into the steamer, while the cook does the main set.
Step 5. Form the racks and add the meat
- Lay the thawed, cured sheets out and fit them together, butting or slightly overlapping the edges, to make a rack of the size you want. Put the striated face down, because that is the show side of the finished rib.
- Trim the exposed fat off the riblet faces that will take the meat, and score the lean. The meat will not stick to fat, because the bond is protein to protein, so any fatty patch is a patch that will not hold.
- Work a little Klebemasse into every seam where two sheets meet, so the sheets bind to one another into a single rack rather than separate strips. The Klebemasse already has its enzyme mixed in.
- Spread a thin, even layer of Klebemasse across the whole top face of the assembled rack, onto exposed lean and not onto fat.
- Spread the reformed meat over the top in an even layer. Make the added part about 60 percent of the finished rib, and up to 70 percent if yield is the priority, with the riblet base making up the rest. The more you add, the more it eats as a formed product and the less as a natural rib, so 60 percent is the balance.
- Press and chill promptly, because the enzyme is already working, and keep fat out of the joins and everything cold throughout.
Step 6. Press and set the green bond cold
Consolidate the rack in a former or under a light press, then hold it at 0 to 4 °C for several hours, and overnight is better. The press brings the sheets and the top layer into firm contact, so the bond forms across the whole face instead of in patches, which is where the hand built first version was weak. A former also holds the higher added water in during the cook. The enzyme matures the bond during the rest and gives the rack the green strength to survive handling into the steamer. Kuraishi and colleagues[5], and the review of Motoki and Seguro[4], describe this cold set behaviour of the enzyme.
Step 7. Steam cook
Steam cook to a safe core temperature, commonly about 72 °C, held briefly. Moist heat suits this product, because it gelatinises the rind collagen and sets the muscle protein into a gel without drying the surface. The heat set muscle gel and the rind gel together become the main binder, while the enzyme, now inactivated above about 70 °C, has already done its job of holding the shape. Chill the rack quickly after the cook.
Step 8. Finish with Origins Smokey BBQ Glaze, then pack
Steam gives no smoke, so the finish supplies it. Origins Smokey BBQ Glaze is a tomato based glaze, simmered to a glossy coat, made with a food grade liquid smoke. The liquid smoke gives the smoke cleanly and consistently, and its phenolic fraction adds a little antimicrobial and antioxidant value. Toss or brush the cooked, chilled racks at about 10 percent of the cooked weight so they are evenly coated, then vacuum pack. The vacuum draws the glaze into the surface and gives a glossy pack.
One safety point must be noted here. The product is now ready to eat, and glazing after the cook reintroduces handling, so the bacteria controlled in the cook can return at the surface. Therefore either hold a strict cold chain with a short, dated shelf life, or give the sealed pack a mild post pack pasteurisation, a gentle in bag heat step that brings the surface up to about 70 °C for long enough to control Listeria. The second route is the safer one for retail. Validate whichever you choose against your food safety plan. Chill and store cold.
Yields and weights
These are planning figures. Confirm them on your own line, because cook loss moves with the cut, the binders and how tightly you press and cook. The water holding work of Hamm[8], of Offer and Knight[9], and the later review of Huff-Lonergan and Lonergan[10], is the reason a phosphated, carrageenan bound, collagen rich product loses little in a gentle cook, so cooking in a former at the low end of the range is how you protect the yield.
A word on where the water sits, because it drives the yield. The 40 percent is water added on the lean of the added part, the primal lean plus the Klebemasse, not on the bone base. Most of it, about 80 percent, sits in the Hautstoss as gel bound water, which is the part that survives the cook[9][10]. In the first version the added part was only one third of the rack, so across the whole rack the added water came to about 9 to 10 percent and the yield was small. The fix is to make the added part most of the rib, about 60 percent, so the water rides on most of the product and the yield rises with it.
Cook loss. Expect about 8 to 12 percent of the green assembled weight in a gentle steam cook to 72 °C, lower toward 5 percent if you cook in a former and the binders are working, higher if you cook the rack loose. Because most of the added water in this build sits in the collagen gel of the Hautstoss rather than free, expect the loss toward the lower end.
A worked example rack, to show the flow of weight, with the added part at about 60 percent and the glaze at 10 percent.
| Stage | Weight |
|---|---|
| Bone base riblets, in | 1000 g |
| Added part, raw, about 60 percent of the green rack | 1500 g |
| Green rack, before cooking | 2500 g |
| Cook loss, about 8 to 12 percent | minus 200 to 300 g |
| Cooked rack, before glaze | about 2200 to 2300 g |
| Glaze added at 10 percent of cooked | about 220 to 230 g |
| Final glazed and packed | about 2420 to 2530 g |
Net weight gain. The lean meat you put in is about 1600 g, that is roughly 650 g of meat on the riblets and about 950 g of primal lean in the added part. The final packed rib at about 2420 to 2530 g is a net gain of about plus 50 percent over that lean. Counted against the whole bone in input of about 2000 g, the gain is about plus 25 percent, because the bone adds weight but takes no water. Either way this is far above the first version, because the added water now sits in the added part, and the added part is most of the rib.
What I learned
A reformed product shows every shortcut once it is cooked and on the shelf, so a bench sample that holds in the hand is not proof, because the real test is whether the join survives a full cook and a customer. Use primals and not trim, cure for colour, bind and safety, bring extracted protein to the join with a Klebemasse, add a rind emulsion for bite and yield, make the added part most of the rib if yield is important, use the enzyme only for green strength, and let a gentle cook set the rack.
References
- Ando H, Adachi M, Umeda K, Matsuura A, Nonaka M, Uchio R, Tanaka H, Motoki M. Purification and characteristics of a novel transglutaminase derived from microorganisms. Agricultural and Biological Chemistry, 1989, 53(10), 2613 to 2617.
- Folk JE. Transglutaminases. Annual Review of Biochemistry, 1980, 49, 517 to 531.
- Kurth L, Rogers PJ. Transglutaminase catalysed crosslinking of myosin to soya protein, casein and gluten. Journal of Food Science, 1984, 49, 573 to 576 and 589.
- Motoki M, Seguro K. Transglutaminase and its use for food processing. Trends in Food Science and Technology, 1998, 9(5), 204 to 210.
- Kuraishi C, Sakamoto J, Yamazaki K, Susa Y, Kuhara C, Soeda T. Production of restructured meat using microbial transglutaminase without salt or cooking. Journal of Food Science, 1997, 62, 488 to 490 and 515.
- Kuraishi C, Yamazaki K, Susa Y. Transglutaminase, its utilization in the food industry. Food Reviews International, 2001, 17, 221 to 246.
- Nielsen GS, Petersen BR, Moller AJ. Impact of salt, phosphate and temperature on the effect of a transglutaminase (factor XIIIa) on the texture of restructured meat. Meat Science, 1995, 41, 293 to 299.
- Hamm R. Functional properties of the myofibrillar system and their measurements. In Muscle as Food, edited by Bechtel PJ, Academic Press, Orlando, 1986, 135 to 199.
- Offer G, Knight P. The structural basis of water holding in meat. Part 1, general principles and water uptake in meat processing. Part 2, drip losses. In Developments in Meat Science 4, edited by Lawrie R, Elsevier Applied Science, London, 1988, 63 to 171 and 173 to 243.
- Huff-Lonergan E, Lonergan SM. Mechanisms of water holding capacity of meat, the role of postmortem biochemical and structural changes. Meat Science, 2005, 71(1), 194 to 204.
- Honikel KO. The use and control of nitrate and nitrite for the processing of meat products. Meat Science, 2008, 78(1 to 2), 68 to 76.