A new paper has been published on the feminizing effect of soy on men:

Messina M. Soybean isoflavone exposure does not have feminizing effects on men: a critical examination of the clinical evidence. Fertil Steril. 2010 May 1;93(7):2095-104. Epub 2010 Apr 8. Review.

You can read the abstract at the link above, and here is a little more info from the June 2010 issue of Soy & Health.

Dear Jack,

I recently got into a debate with a friend who insisted that quinoa, hemp, amaranth, buckwheat, and spirulina are all complete proteins. When I told him soy is the only complete vegetarian protein, he told me he avoids all soy due to its negative effect on men’s health. I wondered where this guy learned this and a quick google search on vegetarian complete proteins got me here. I checked out your page on protein. That chart listing the essential amino acids in so many plant foods made me realize that in fact many plant foods do contain all the essential amino acids but at such low levels it’s hard to list them as ‘complete proteins’ in the true meaning intended. So my take is that the only vegetarian protein that comes close to really being a complete protein (meaning it has essential amino acid levels similar to those of animal proteins) is soy. Would you agree?

Answer:

[As an aside, the research to date indicates that moderate amounts of soy do not harm men's health.]

Because this question about complete proteins comes up on a regular basis, I decided to do some number crunching to get to the bottom of it. Mind you, there are many more numbers to crunch if someone wanted to take the time, but I think I’ve covered enough to draw some conclusions.

I see two ways of looking at the question of whether a food has complete protein:

If you look at Table 3 of Where Do You Get Your Protein?, you will see what it takes to meet the RDA for each EAA for a variety of foods.

I have taken some of the data from Table 3 and constructed the table below to look more closely at what foods might contain a “complete protein”. In the table below, the numbers in the right-hand column represents the following equation:

Where:

From the chart above, it appears that tuna is the most complete protein with a range of only .4 servings and only 1.7 servings required to meet the RDA for all the EAA. Chicken, beef, edamame (whole, cooked soybeans), lentils, and pinto beans all do quite well. I think it’s fair to consider all of them a “complete protein”. Tofu, milk, quinoa, soy milk, and eggs do significantly better than most grains and nuts which have a much wider range.

As for spirulina…not so much.

One thing to note about this is that all the numbers depend on the serving size. I tried to pick what I thought were reasonable (or common) serving sizes for each food (you can see what they are in Table 3).

The USDA database has a lot of specific entries for some of the food categories above (like ground beef). I chose what looked like a common version of the food, but I did not average the data across more than one version.

As for the remaining, supposed complete proteins mentioned in the original question above, here is what I found:

Amaranth
The USDA lists 9 g of protein per cup of cooked amaranth. That’s a good amount when compared to other grains, but I’m not sure if it’s as easy to eat a cup of amaranth as a cup of, say, rice or corn. The USDA had no amino acid info.

Hemp
There was no info in the USDA database. I found many sources saying that it is a complete protein but none that I know to be reliable.

Nutritional yeast
No info in the USDA database. According to this site, it has 9 g of protein per 3 tbsp serving. That’s a decent amount, but I’m not sure if we can trust that info.

In conclusion, edamame, lentils, and pinto beans fared pretty well with chicken and beef for being a complete protein. Tofu, soymilk, and quinoa were on par with eggs and milk.

A month old now, but I finally got around to sending out a link to this article Ginny Messina wrote, Hexane in veggie burgers: little science behind the claims.

Article from Men’s Fitness. Link.

Thanks, Matt!

I just updated Another Internet Soy Article with the following:

A 2009 report from the Shanghai Breast Cancer Survival Study showed that women with a prior diagnosis of breast cancer (including estrogen-positive), who ate more soy, had lower rates of death and cancer recurrence (12). The study followed women for an average of 3.9 years after a diagnosis of breast cancer. The researchers measured a beneficial effect of up to 11 grams of soy protein per day. Table 3 shows the results.

12. Shu XO, Zheng Y, Cai H, Gu K, Chen Z, Zheng W, Lu W. Soy food intake and breast cancer survival. JAMA. 2009 Dec 9;302(22):2437-43.

Thanks to the Vegetarian Society of Hawaii, you can now watch a video of my talk Vegan Nutrition: What Does the Science Say on-line. Click here.

A special thank you to Dr. William Harris for filming and putting the video together!

A study was just published showing that Chinese women living in Singapore who ate more soy had a reduced risk of hip fractures. Compared with women in the lowest one-fourth of intakes for tofu, soy protein, and isoflavones, those in all three of the higher intake categories had a 21%–36% reduction in risk. Soy did not show any benefit for men.

Koh WP, Wu AH, Wang R, Ang LW, Heng D, Yuan JM, Yu MC. Gender-specific associations between soy and risk of hip fracture in the Singapore Chinese Health Study. Am J Epidemiol. 2009 Oct 1;170(7):901-9. Epub 2009 Aug 31.

I just updated VeganHealth.org with information from an April 2009 review on soy infant formulas (link). Here is an excerpt from the paper:

“Even though soy isoflavones can bind and activate [estrogen receptors (ER)], they do not behave like typical estrogen agonists but rather as selective ER modulators and, in addition, have many other actions that are ER independent, eg, tyrosine kinase inhibition. It is unfortunate that soy isoflavones have been called “phytoestrogens,” because they are not estrogens and are not truly estrogenic at nutritionally relevant concentrations. The weak isoflavone potency for activating the ERs combined with competition with endogenous estrogens for the ERs make isoflavone-related ER activity minimal when fed in amounts similar to those found in [soy formula], even when fed during early development. Moreover, although some studies have shown similar gene expression profiles for genistein (the major soy isoflavone) and [estrogen] in some tissues in vitro and in vivo, ingestion of soy foods results in a complex mixture containing hundreds of phytochemicals and peptides being introduced to the gastrointestinal tract, many of which are absorbed and have biological actions. This situation is not unlike the mixture of phytochemicals found in a typical meal containing a mixed salad and vegetables.”

I just updated Another Internet Soy Article on VeganHealth.org with information on breast cancer. Link. It’s too much to reprint here but it is all good news, with decent evidence that eating soy in moderate amounts can actually decrease the risk of breast cancer. I wouldn’t go so far as to claim that we know it does, but at the very least we can conclude that it doesn’t increase the risk for breast cancer.

Ginny Messina has sent me an update to the article, Is It Safe to Eat Soy?, which she and her husband Mark wrote for VeganHealth.org.

Quick summary: Soy does not appear to lower sperm count or testosterone levels.

More details can be found here.