Spinach and Ricotta Cannelloni

What this is

Pasta tubes filled with spinach and ricotta, baked under tomato sauce and a little cheese. An hour, six servings, and it feeds people who claim not to like vegetables.

It is also a dish where the two main ingredients each have a genuinely interesting story behind them, one about waste and one about a compound that gets talked about badly.

The method is written for dried cannelloni tubes, which need more sauce than people expect.

Ingredients

For 6 servings

Filling 500 g ricotta 500 g spinach, fresh or frozen 80 g parmesan, grated 1 egg 2 cloves garlic, minced Grating of nutmeg Zest of half a lemon Salt and black pepper

Sauce 2 tins chopped tomatoes 1 onion, finely diced 3 cloves garlic, sliced 3 tbsp olive oil 1 tsp dried oregano 1 tsp sugar Salt and pepper

To assemble 250 g dried cannelloni tubes, about 18 125 g mozzarella, torn 30 g parmesan Basil

Method

1. Heat the oven to 190C.

2. Wilt the spinach, then squeeze it genuinely dry in a clean cloth. Wet spinach makes a watery filling and a soggy bake, and this is the step that decides the dish.

3. Chop it and mix with the ricotta, parmesan, egg, garlic, nutmeg, lemon zest and plenty of seasoning. Ricotta is bland; it needs more salt than you think.

4. For the sauce, cook the onion in the oil 10 minutes, add garlic and oregano 1 minute, then the tomatoes, sugar and a splash of water. Simmer 15 minutes. Keep it loose; the pasta will drink it.

5. Spread a third of the sauce in a baking dish.

6. Pipe the filling into the tubes using a piping bag or a freezer bag with the corner cut off. A teaspoon works and takes three times as long.

7. Lay them in the dish, cover completely with the rest of the sauce, then the mozzarella and parmesan.

8. Cover with foil, bake 35 minutes, then uncover and bake 15 more until browned. Rest 10 minutes before cutting.

Ricotta is the by-product, and that is the point

Most people do not know what ricotta is. It is made from whey, the liquid left after cheese curds are separated, and there is a study that maps exactly where everything goes.

Researchers evaluated the partitioning of milk components during traditional pasta filata cheesemaking through closed mass balances, across three cheesemaking plants over a full year, sampling during bimonthly trials and determining the distribution of water, proteins, fat, ash and lactose.

A year, three plants, bimonthly sampling. And the mass balances closed at greater than 98.7 percent for every component, which means almost nothing was unaccounted for.

Average recovery in the cheese itself was 72.0 percent for proteins and 76.9 percent for fat, while lactose remained predominantly in the whey at 86.6 percent.

So roughly a quarter of the protein and the fat leaves with the whey, along with nearly all the milk sugar. That is what ricotta is made from.

The authors state it directly: integrated ricotta production significantly contributed to the overall recovery of nutrients at the system level.

Which is a rare thing to be able to say about a food. Ricotta exists because somebody looked at a tank of by-product and worked out how to get the remaining protein back out of it, and the practice is centuries old.

This study concerned Caciocavallo Palermitano, a specific traditional cheese, and measured component partitioning and yield prediction rather than nutrition or health. Commercial ricotta varies widely, and some sold as ricotta is made from whole milk rather than whey.

The spinach, and being honest about oxalate

Spinach carries oxalate, and the internet handles this badly in both directions.

A narrative review examined oxalate as an emerging contributor to cardiovascular disease. It describes oxalate as a small molecule found in plant-derived foods such as spinach, potatoes, almonds and peanuts, which is also produced by the body itself.

That last clause matters and is usually left out. Your body makes oxalate whether or not you eat spinach.

The review notes that oxalate is traditionally studied in the context of kidney stone disease, and that recent evidence suggests it may be a dietary contributor to inflammation and oxidative stress in cardiovascular disease, with elevated systemic levels promoting reactive oxygen species generation and activating inflammatory pathways including NF-kB, MAPK and the NLRP3 inflammasome.

Those are mechanisms, and mechanisms are the weakest form of evidence about what to eat. The authors themselves use the word emerging, and frame it as evaluating emerging evidence rather than reporting an established risk.

So the honest position: this is an area of active investigation, not a reason for a healthy person to stop eating spinach. If you have had kidney stones, oxalate is a real conversation, and it is one to have with your doctor rather than with an article.

One practical note that is genuinely useful: oxalate is water-soluble, so wilting spinach and squeezing out the liquid, which step 2 tells you to do for texture reasons, also removes some of it.

This is a narrative review evaluating emerging mechanistic evidence, not a trial showing that dietary oxalate causes cardiovascular disease. Nothing here is a reason to change your diet, and kidney stone history belongs with a clinician.

How well we actually know what dairy does in digestion

A methodological review worth knowing about, because it tells you how much of the dairy nutrition literature is built on a model rather than on people.

The INFOGEST protocol is a standardised in vitro digestion model used to evaluate digestibility and bioaccessibility of nutrients. This review analysed 50 studies applying the updated INFOGEST 2.0 model to milk and dairy products since 2020.

Fifty studies, all using a simulated digestive tract in a laboratory.

The review found the model performs strongly in reproducing general digestive trends, including peptide fingerprint profiling, consistently high protein digestibility, and matrix-dependent lipid and mineral bioaccessibility, and is particularly effective at evaluating how structural modifications to dairy products change digestive behaviour.

Matrix-dependent is the interesting phrase. The same nutrient behaves differently depending on the food it arrives in, which is why cheese and milk are not interchangeable even at identical composition.

The limitations the authors list are the reason to cite this at all: application remains skewed toward bovine systems, methodological variability and protocol modifications pose challenges to reproducibility, and reliance on simplified downstream models constrains physiological interpretation of bioactivity and nutrient absorption.

A review that spends its conclusion explaining why its field's results should be held loosely is doing the reader a service.

This is a review of an in vitro digestion protocol, not a study of ricotta or of this dish. It is cited for what it says about the strength of the underlying method, which applies to a great deal of the dairy nutrition claims you will read elsewhere.

Storage and swaps

Keeps 3 days in the fridge and reheats better than most baked pasta, because the sauce keeps it moist.

Assemble ahead and refrigerate unbaked for up to 24 hours. Add 10 minutes to the covered bake.

Freezes 3 months assembled and unbaked. Bake from frozen, covered, at 180C for about 70 minutes, then uncovered.

Swap cannelloni tubes for fresh lasagne sheets rolled around the filling, which is easier than piping and looks better.

Swap spinach for chard, kale or roasted squash. All need the same squeezing-dry treatment except the squash.

Add a layer of bechamel over the top if you want it richer, though the sauce and mozzarella do the job.

And squeeze the spinach harder than feels reasonable. Every failed version of this dish is a wet one.