This article first appeared in the St. Louis Beacon, May 27, 2009 - Early this year, the Food and Drug Administration approved the first drug produced by a genetically engineered farm animal. In what I am sure will be followed by many pharmaceuticals produced on the farm ("pharming" ... get it?), a human gene was transferred into goats and the drug harvested from the goat milk.
As you might suspect, the drug was not a simple chemical like aspirin that can be produced cheaply in bulk. It was a protein, a big molecule with long chains of atoms wound up in complex ways to produce a particular 3-D shape that must be just right for the protein to work correctly in the body.
Protein drugs have always been a problem for the pharmaceutical industry to produce. Before the advent of genetic engineering, proteins such as insulin had to be purified from the body parts of pigs or cows, which contain a version of the protein similar to the human one. As you can imagine, it wasn't cheap to produce protein drugs this way.
Genetic Engineering
Genetic engineering provided a vastly easier approach. Instead of attempting to purify the sought-after protein from limited amounts available in difficult-to-preserve animal tissue, the human gene encoding the protein is put into a bacterial cell, the cell allowed to multiply into trillions of cells in a culture vat, and the cells burst to harvest the protein.
Early successful attempts were really just that simple. With a little tinkering, they worked quite well. The most important trick bioengineers learned was to embed the human gene within a bacterial gene sequence that was read often by the bacterial protein-making machinery - in essence, adding signals to the front of the gene that commanded the bacterial cell to continually make the protein the gene encodes.
While this "stick the gene into bacteria" approach worked well for producing simple proteins like insulin, more complicated proteins presented too much of a challenge for bacterial cells. The process of making proteins is far more sophisticated in a eukaryotic cell (a cell with a nucleus), and bacterial cells simply lack the necessary equipment.
What to do? Animals like us -- as well as plants, fungi, slime molds and algae -- are all eukaryotes. Why not harvest the human gene from cultures of a eukaryotic cell? Good idea, and it works great.
Many of the protein drugs you read about in the newspapers (or, more trendily, on a web-based site like this one) -- cancer drugs like Avastin and arthritis drugs like Enbrel -- are produced in huge stainless steel vats of genetically engineered Chinese hamster ovary cells. The cells grow well in culture and churn out large amounts of the desired proteins in fully active form.
However, a cell culture factory costs hundreds of millions of dollars to build, and so only drugs with a very large potential market are candidates, however pretty the science of the process. Protein drugs used to treat rare disorders would never be used often enough to justify the huge expense of cell culture production.
Hence the goats. While a cell culture factory costs hundreds of million of dollars, a genetically engineered goat costs thousands. A herd of 200 genetically engineered goats living under carefully controlled conditions can be produced and operated for a few million dollars.
In essence, you are substituting a goat for a stainless steel culture vat. Because it costs so much less to produce the protein in a goat than in a vat, "goat pharming" permits the commercial production of drugs with a small potential market.
The drug approved last February by the FDA was a human anticlotting protein called antithrombin that had until now been purified from human plasma donations. It has often been in short supply or unavailable because of blood donation shortfalls. One goat can produce as much antithrombin in a year as can be obtained from 90,000 blood donations.
It's hard to miss the point: Pharming lowers medical costs.
Taking lessons from the pioneering bacterial approaches, the goat pharmers used genetic engineering to transfer the human antithrombin gene into goat DNA so it was linked to a goat DNA sequence that normally controls production of a protein found in milk.
Presto! The goat's milk is rich in antithrombin, and this anticlotting protein is not made anywhere else in the goat.
How Did They Manage This?
The goat DNA was attached to the purified human gene, and this combo was injected into a one-cell goat embryo (actually, they took one cell from an eight-cell embryo, a routine procedure in animal breeding these days). The injected cell was then transplanted into a surrogate goat mother. After birth and rearing, the goat (if female) produces antithrombin in its milk.
The beautiful thing about all this is that you only have to do the genetic engineering a few times. Once you have a boy goat and a girl goat, you have only to let nature take its course and soon you have a herd of goats all producing buckets of antithrombin.
The FDA-approved goat pharm that is making antithrombin is located in Framingham, a small town in central Massachusetts. Similar operations are sprouting up in other countries.
The Dutch company aptly named Pharming is requesting FDA approval of a protein drug to treat angioendema, a protein deficiency that leads to swelling of body tissues. How are they producing the protein drug? They are harvesting it from the milk of transgenic rabbits!
And goats and rabbits are only the beginning.
Even Moss
This month German researchers reported that they have succeeded in harvesting human proteins from genetically modified moss! Like many long-shot experiments that no sensible scientist thinks will work, this one was carried out by a graduate student, Marc Gitzinger, a PhD student at the University of Freiburg.
Plant bioengineers have long known that if you move a human gene into a plant, the corresponding human protein is not produced - the gene reading and protein making machinery of a plant is just too different from ours. But Gitzinger had not yet gained the experience of his mentors, and so set out to try anyway, reasoning that a very primitive plant might not yet have evolved all the plant differences. He chose a moss, Physcomitrella patens, one of the most primitive of plants (it evolved 450 million years ago, long before seeds and flowers), and simply inserted unmodified human and mammalian genes into moss cells.
I bet no one was more surprised than his professor when the genetically modified moss started manufacturing the human proteins!
Gitzinger's experiment is going to be reproduced a lot in coming months, as it suggests that pharming can be carried out cheaply indeed. All you have to give genetically modified moss is water, a couple of nutrient salts and a little light. That's even cheaper than keeping a goat - and you don't have to milk it.
George B. Johnson's "On Science" column looks at scientific issues and explains them in an accessible manner.
Johnson, Ph.D., professor emeritus of Biology at Washington University, has taught biology and genetics to undergraduates for more than 30 years. Also professor of genetics at Washington University’s School of Medicine, Johnson is a student of population genetics and evolution, renowned for his pioneering studies of genetic variability. He has authored more than 50 scientific publications and seven texts.
As the founding director of The Living World, the education center at the St Louis Zoo, from 1987 to 1990, he was responsible for developing innovative high-tech exhibits and new educational programs.
Copyright George Johnson