Adaptive Immunity: How Randomness Comes to the Rescue

| By on Endless Forms Most Beautiful

Note: Originally posted May 13, 2010.

The last time my husband and I went camping in the desert, we noticed something: it’s a dangerous place! Everything seems to be trying to kill you. Besides extreme temperatures and little green men (which we didn’t see, sadly), you have to watch out for things like snakes, scorpions, spiders, ticks, cacti, and the rare but devastating hantavirus.

Even in the relative safety of your home, viruses, bacteria, parasites, and fungi are an inescapable part of life. Without a functioning immune system, we would easily succumb to infection from these pathogens. All animals share a basic set of defense strategies called the innate immune system, which includes protective barriers like skin and special cells that gobble up foreign material. In addition to this first line of defense, vertebrates have a highly sophisticated secondary system in place, called the adaptive immune system.

Previously, we examined the role of randomness in self-assembly processes and chaotic systems (here and here). Today I want to explain how the adaptive immune system harnesses the power of randomness to protect the body from assaults it has never seen before. To me, this topic represents a compelling example of the way God works through natural processes that he has put in place to preserve and uphold life.

Adaptive immune responses come in two types: antibody responses and cell-mediated responses. Each type is executed by a different kind of white blood cell, or lymphocyte. B cells, produced in the bone marrow, generate antibodies, while T cells, produced in the thymus, directly or indirectly kill pathogen-infected cells. Here we will focus on how antibodies are made within B cells, as their production requires randomness at multiple levels.

Antibodies: Signposts for Destruction

The main function of an antibody is to bind tightly and specifically to an antigen, usually an exposed portion of a virus or bacterium. This binding can inhibit the normal function of the pathogen, just as your ability to walk is inhibited when a toddler hugs you around both legs. Antibody binding also acts as a signpost, alerting cells of the innate immune system to destroy the pathogen.

Antibodies are proteins. Like all proteins, antibodies are strings of amino acids that fold up into a particular 3D shape. As you can see from the illustration at left, antibodies are Y-shaped proteins made from four separate amino acid chains: two “light” chains and two “heavy” chains. (The heavy chains are about twice as long as the light chains, so they have a higher molecular weight.) All antibodies are identical in the blue “constant” region but are quite different from each another in the orange “variable” region. The shape of the variable region determines what the antibody will bind to (if anything) and is consequently called the antigen-binding site.

If the antibody were a tool in your garage, it might be an 8-in-1 screwdriver, with the constant region being the handle and the variable region being an exchangeable bit. Each bit is designed to work with a different kind of screw, and as you probably know, using the wrong size or shape of bit can be fruitless and frustrating. Similarly, the shape of the antibody’s variable region must be exactly complementary to the antigen for it to bind.

Mix-and-Match Mutations Create Diversity

Amazingly, animals can manufacture antibodies for not just eight but literally trillions of foreign particles, even synthetic compounds never encountered in nature. How can a B cell design an antibody for a novel antigen when its shape depends on the amino acid sequence, which, in turn, is encoded by a gene? Humans only have about 25,000 genes total, so it can’t be that each potential antibody is generated by a separate gene.

Surely the design of a highly specific antigen-binding site could not be accomplished by random mutations! Actually, yes, though as we will see, the story is not as simple as a single gene undergoing spontaneous mutations every now and then. The mutations are programmed to occur in just a tiny part of the genome, and instead of involving changes in single DNA “letters” (bases), they involve random rearrangements of whole gene segments.

I’ll explain by way of analogy. Imagine packing for a trip. You will be gone for two weeks, but because your airline charges ridiculous fees to check luggage, you opt to take only a small duffle bag. You have a wide variety of events planned, so each day will require a different outfit. Since space is limited, you grab your favorite pair of shoes—they go with everything (work with me, ladies; analogies aren’t perfect)—along with three pairs of pants and five tops. Assuming each top matches each pair of pants, you can make 15 unique outfits. Some are more casual while others are dressier; some are better for hot weather while others are more appropriate for rain. In other words, by including a variety of styles, you can cover all your fashion needs with great efficiency.

Amazingly, B cells are able to produce a huge variety of antibodies by mixing and matching DNA segments to create unique antigen-binding sites. The pair of shoes is like the constant region; it is used with every “outfit.” Each pair of pants is like a different heavy chain, while each top is like a different light chain. When a new B cell begins to mature, it randomly picks some combination of gene segments (an outfit) from many options in the genome (the suitcase). It then physically stitches the segments together and from now on only produces one kind of antibody. (That’s like only getting to wear one outfit for the rest of your life!)

I’ve dramatically oversimplified the picture here. In reality, to make the variable region for its heavy chain, the maturing B cell randomly chooses one possible segment from each of three segment types, called V, D, and J (for variable, diversity, and joining, respectively). There are 51 V segments to choose from, 6 J segments, and 27 D segments, which yield 8,262 (51x6x27) possible heavy-chain variable regions. The light chains are a bit simpler, as they don’t use D segments. There are 316 possible combinations of V and J segments for the light chain. Thus, there are 2.6 million (8,262x316) different antigen-binding sites possible from this mix-and-match process.

More Diversity from Built-in Imprecision

2.6 million is a lot, but not even close to the trillions of antigens out in the world. Where does the rest of the diversity come from? As it turns out, the process of stitching together the V, D, and J gene segments into light and heavy chains isn’t very precise. To recombine (reshuffle) the gene segments, two lymphocyte-specific proteins, RAG1 and RAG2 (for recombination activating genes), form a complex to first physically break the double-stranded DNA in specific (but unpredictable) places. Then the complex works together with a cleanup crew of other proteins to rejoin the DNA segments in a different order.

Normally when DNA breaks occur, say due to radiation from the sun, this same kind of DNA repair occurs to rejoin the two ends very precisely. In developing B cells, however, a few bases can be added or removed, analogous to lengthening the hem of your pants or cutting the sleeves off your shirt. This built-in imprecision leads to an estimated 100 million-fold increase in the diversity of antigen-binding sites.

This boost in diversity comes at a significant cost, however. Since each amino acid is encoded by three DNA bases, the addition or deletion of only one or two bases will cause a shift in the “reading frame” such that the DNA code becomes meaningless and the B cell cannot make a functional antibody. Many developing B cells suffer this fate and die in the bone marrow, never getting to debut their one outfit. Yet, the cost is worth it to the organism for the benefits that come with a robust immune response.

What’s the Point?

Today we have seen two ways in which B cells exploit the power of randomness to make an enormous repertoire of antibodies. First, the RAG proteins break and rejoin the DNA at random sites to create unique combinations of V, D, and J gene segments. Second, a random small number of bases can be added or lost when the DNA ends are rejoined, leading to the insertion or deletion of one or more amino acids in the antigen-binding site.

The point of my post today is not to reveal the ingeniousness of the adaptive immune system, though I hope it accomplishes that also. Rather, I want to emphasize that God uses natural processes—indeed, even a “blind” system for generating massive amounts of diversity—to carry out his purposes. If God uses natural mechanisms that work over short time scales (less than a week) to evolve life-giving solutions to disease, mightn’t he also use a similarly elegant approach to creating life over long periods of time?

In my next post, we will continue to see how B cells and antibodies function and begin to see if there is any evidence for evolution of the immune system.




Applegate, Kathryn. "Adaptive Immunity: How Randomness Comes to the Rescue" N.p., 22 Mar. 2013. Web. 23 February 2017.


Applegate, K. (2013, March 22). Adaptive Immunity: How Randomness Comes to the Rescue
Retrieved February 23, 2017, from

References & Credits

Alberts, Bruce et al. “The Adaptive Immune System.” Molecular Biology of the Cell. Fourth Edition. 2002.

Story, Craig M. “The God of Christiantiy and the G.O.D. of Immunology: Chance, Complexity, and God’s Action in Nature.” Perspectives on Science and Christian Faith. 61:4. Dec. 2009.

About the Author

Kathryn Applegate

Kathryn Applegate is Resources Editor at BioLogos. She received her PhD in computational cell biology at The Scripps Research Institute in La Jolla, Calif. At Scripps, she developed computer vision tools for analyzing the cell's infrastructure, the cytoskeleton. Kathryn joined the BioLogos staff in 2010.

More posts by Kathryn Applegate