Comments on Kathryn Applegate’s May Posts on BioLogos
|October 7, 2010||Posted by Caroline Crocker under Intelligent Design|
Since I am a cell biologist and immunologist by training, it is with great interest that I read Kathryn Applegate’s May BioLogos posts drawing parallels between adaptive immunity and evolution. In the first essay she claims that antibody “production requires randomness at multiple levels” and that God may use random processes to create “life over long periods of time.” In the second post Dr. Applegate goes on to suggest that evolution uses “the same kinds of mechanisms, except the mutations occur in germ cells…”
These are interesting hypotheses, but I am not convinced that the elegant processes whereby B cells differentiate and germ cells are formed actually give rise to the conclusions drawn. Good science is dependent on accurately distinguishing between data, interpretation of data, extrapolation from data, and even speculation; in these posts this has not been adequately accomplished. In fact, even the science is faulty in places. To explain, the data shows that B cells manufacture over 1015 different antibodies using less than a couple of hundred gene segments. They accomplish this feat by rearrangements and excision of DNA sequences—these occur in a highly regulated fashion that has been extensively described in the literature. These facts have been established by interpretation of vast amounts of data.
However, I would like to suggest that the claims that B cell differentiation is 1) random, 2) a model for the way that God created life, and 3) that evolution “works” by B-cell-like mutations in the germ cell line, or 4) that germ cell formation is in any way analogous to antibody formation are based on a one-sided explanation of the science and much speculation. Dr. Applegate states that God could have done it this way; I do not dispute this. After all, if He is God, it is logical that He can do whatever He wants.
Unfortunately, in science the question is not what God could do, but what actually happens/ happened or, in essence, what He did do. In fact, unless we cross the line from interpretation to speculation, much further research is required before any conclusions about the parallels between the B cell picture, germ cell formation, and macroevolution can be drawn.
Let’s look at each of the above in turn, assessing the scientific merit of the claims. Is the process whereby stem cells become fully functioning, specific antibody-producing B cells random? No. In fact, it is anything but random and much too complex and intricate to describe here. Lippincott’s Immunology and indeed Applegate herself describe the process as elegant. In elucidation of this question, instead of glossing over the process and majoring on the randomness, it would be beneficial to examine the sequence of events in at least a little more detail.
Starting with the basics, the prime function of a B cell is to make immunoglobulins (antibodies); the base structure of an antibody consists of one heavy and one light chain of protein (See http://8e.devbio.com/article.php?id=31 for a nice image). These chains are encoded in DNA, which is much like an instruction book for the cell. The DNA is located in the nucleus of a cell. When some of the information in the DNA is needed, that part of the DNA is transcribed into messenger RNA (mRNA), which carries the information into the cytoplasm of a cell. There, complex machines translate the information in the mRNA into specific proteins—antibodies or immunoglobulins are a specific type of protein. A human being’s DNA instruction book is divided into 23 chromosomes or volumes (more about this later).
All DNA coding for the immunoglobulin heavy chain is found on chromosome 14: this chromosome has DNA encoding several types of immunoglobulin C (constant) regions, 27 types of D (diversity) regions, 6 types of J (joining) regions, and 65 (not 51 as Dr. Applegate claims) V (variable) regions. The finished immunoglobulin heavy chain will then consist of three or four C, one D, one J and one V region. And, as the article stresses, which D, J and V region are used to make the protein is random. However, it is important to realize that the process whereby this is accomplished is extremely controlled and accurate—in other words, the mixing-up is deliberate. First the DNA between a randomly-selected D region and a randomly-selected J region is deleted, joining these two together. Then the process is repeated to join a V region to the D region, making a VDJ gene. The steps to put these DNA segments together are orchestrated by protein machines so that they are carried out precisely and sequentially. In addition, as the cell successfully completes each step, it puts signals on its surface to alert the nearby cells where it is in the process. Finally, the VDJ DNA and DNA from the C region (δ and μ, to be specific) is transcribed into one long RNA, the intervening sequences removed, and the RNA processed to maturity.
After translation of this mRNA into protein so that a particular type of immunoglobulin (IgM) can be found in the cytoplasm, the cell begins light chain construction—signaling the fact on its surface by display of at least two markers (IL-7 receptor and CD19). So, the cell is not just carefully constructing antibody in a time and place-controlled manner, it is also communicating where it is in the process to the surrounding milieu. (As the cells progress through the stages until they become fully functional B cells, immunologists give them different names, like pro B cells, pre B cells, and more—I will just call them all B cells to avoid confusion.)
Light chains consist of one of two C (constant) regions, a V (variable) region and a J (joining) region. The DNA sequences for light chains are found on chromosomes 2 and 22. There are approximately 100 V variants, 4 to 6 J variants, and two types of C region (κ or λ). Again, construction of a light chain is highly regulated. First, the DNA for one of the V and one of the J regions is spliced together and the intervening DNA discarded. Next, this DNA and the DNA encoding one of the C regions is transcribed into one long RNA molecule. The cell edits out the intervening RNA and a mature mRNA molecule consisting of VJC is formed. This mRNA is translated into protein and the light chain is formed.
Note that the above paragraph says that a V region and a J region are spliced together, but this is not as simple as it sounds. For example, one might ask how the B cell “knows” where one “V” section ends and another “V” starts and what exact part of the DNA should be spliced to the “J” section. After all, the DNA code has only four letters (A, C, T, G) and one part of DNA must look very similar to another. This is regulated by special RSS sequences that can be found at the ends of each V and each J region. A group of protein machines (enzymes) called recombinases recognize the RSS sequences at the ends of each region, and another group of enzymes (RAG) cut the DNA in those places. The ends are deliberately modified by an enzyme terminal deoxytidyl transferase which adds random “letters,” and the DNA is spliced back together by still another type of enzyme called a ligase. That way most heavy chains only have three or four C’s, one D, one J, and one V. Ingeniously, sometimes the regions are put in backwards and sometimes more than one region is utilized, further increasing the diversity or possible antibodies generated. In addition, some special regions of the DNA can and do undergo point changes—again to increase the diversity. However, note that this is only random in the sequence generated—it is very specific as to the particular places and the exact times of occurrence.
Next, the antibody-displaying B cells are tested in various ways, such as whether they recognize self. Those that are found unsuitable commit cellular suicide, another highly-regulated and precise process called apoptosis that has the goal of not spilling cellular “guts” and making a mess! The rest of the B cells are allowed to leave the bone marrow for the next round of maturation—also an exceedingly controlled and information-rich process. (Dr. Applegate’s claim that many cells die because they have a frameshift mutation is inaccurate.) Amazingly, this tremendously simplified explanation of the process of antibody formation only outlines what occurs before the B cells encounter antigen (a “foreign” substance)! My extrapolation from this information is that the process looks deliberately engineered for generating a diversity of antibodies rather than something that could be used as evidence for the mechanism of evolution because it displays the “power of randomness”.
This brings us to the question of whether “the generation of antibody diversity” is an example of how “God creates and sustains life”. Dr. Applegate asserts that because God uses this supposedly “blind system,” He uses the same process in evolution, so that mutations and natural selection give rise to the diversity of life. This is akin to saying that the careful merging of the parts of a manuscript, putting a table of contents at the beginning, the introduction next, the body in the middle, and references at the end (analogous to the B cell generation process), making sure that only one version of each is included, is the same as altering the manuscript by typos. Having just released a book (http://www.freetothink.us), I can tell you that this is definitely not the case!
Diversity of antibodies generated by B cells is due to deliberate, cell-engineered changes in the DNA sequence, not random mutations. In fact, I have never before heard the process whereby functional antibodies are formed (before they encounter antigen) described as mutation. And it is well-known that the appearance of functionality as a result of a mistake-mutation is extremely rare. Of course, after encountering antigen the hypervariable regions of the antibody DNA do undergo somatic hypermutation, but again this is in particular places and is controlled by enzymes. Using some of the examples Dr. Applegate cites, loss or gain of an entire somatic chromosome in a human results in miscarriage or, in the case of our smallest chromosome, 21, Down ’s syndrome. Duplication of the entire genome only occurs in plants and does not appear to confer a significant advantage. The literature reveals that insertions, duplications, point mutations, and inversions cause a myriad of genetic diseases. Perhaps to assert that B cell production of novel antibodies is a picture of the way God creates and sustains life through random mutation and natural selection is speculation and even invalid extrapolation from the science.
Now, let’s consider whether the generation of diversity in the germ cells (sperm and egg) is at all analogous to what occurs in the B cell. Meiosis, which is the process whereby germ cells are formed, is like a highly-regulated dance performed for the purpose of mixing up the genetic material (See second image at http://avonapbio.pbworks.com/Chapter-13). In order to understand this process, one must first know that every cell of a human has two copies of every one of their 23 chromosomes: one copy comes from their mom and one from their dad, making a total of 46. Put simply, genes are subsections of chromosomes, like sentences are subsections of books. Each of the 23 chromosomes code for completely different genes, but the copies from mom or dad only vary by the specific nature of the gene. Take a silly example. Say if chromosome 2 codes for nose shape; then the copy of chromosome 2 from mom (I will call it 2“m”) might code for hooked nose and the one from dad (2d) for snub nose. Chromosome 2 contains a gene that codes for nose shape, but the particular shape is dependent on the specific version of the gene.
A baby is a result of the fusion of a sperm and an egg, but of course the baby, being human, also has 46 chromosomes in each cell. Therefore, the germ cells that fuse during conception can only have 23 chromosomes each. Meiosis is the process whereby the 46 chromosomes in the progenitor cells are pared down to 23 in the germ cell. Whether each chromosome (1-23) is the version from mom (m) or the version from dad (d) is random, but the process has to be highly regulated because each germ cell must only have one copy of each chromosome—duplicate somatic chromosomes or deletion of an entire somatic chromosome result in miscarriage. (One can get away with more than or less than two sex chromosomes.) The variability in the genetic nature of the germ cells is further enhanced by a process that occurs during mitosis where two chromosome 1’s or two chromosome 2’s etc. can swap tips. The resultant chromosome is an amalgam of the originals—but again note that 2 can only swap with 2 and 3 with 3, etc. Other swaps (e.g. between 2 and 5) cause genetic disease. Again, the process is highly regulated and there are many mechanisms in place to minimize the chances of occurrence of a mutation.
So, how is this all accomplished? Well, the process is nothing like the way B cell diversity is generated. First, all the chromosomes in the progenitor cell are copied; now the cell has four copies of each chromosome making 92 total (the chromosomes are now given a different name, but we will skip that for the sake of clarity). In our example above, the cell now contains two 1m and two 1d, two 2m and two 2d, etc. up to two 23m and two 23d. 1m and 1d are referred to as homologues and 1m and 1m are called sisters. Next, in a step that takes most of the time given to meiosis, the sisters and homologues are organized into groups of four: all the 1’s are stuck together, all the 2’s, all the 3’s, etc. This is when the tips of homologues can crossover and be swapped, resulting in chromosomes with, perhaps, a body that is 2m and tips that are 2d. The scientific name for this process is synapsis and chiasmata formation and numerous papers have been published describing the complex machinery needed.
The tetrads are then attached to cellular filaments or microtubules, lined up along the middle of the cell, and pulled apart so that the homologues progress to different poles of the cell and the sisters to the same. However, which homologue goes to which side is random. Next, the cell divides, so each new cell has 46 chromosomes worth of DNA, but only 23 chromosomes worth of information (remember that the sisters, which are identical, except for maybe their swapped tips) go to the same pole). But, the two daughter cells have different versions of each chromosome. These cells then divide again, going through a similar dance, and the germ cells, which each have 23 chromosomes) are formed. It is possible that mutations occur during this process and these are passed on to the offspring, but note that there is NO deliberate excision and splicing of genetic material, no production of antibodies, in fact virtually no similarity between this process and that of B cell antibody production. In fact, the only similarity between these processes is that both are marvels of precise engineering and nanotechnology.
Now, it may be that the mechanism of evolution is random mutation followed by natural selection; this is a theory accepted by many scientists. But, it is vital to remember that scientific facts are not decided by popularity, but by data and interpretation. So, in order to evaluate the theory, let’s stick with the science, not with speculation and insufficiently substantiated conclusions based on an inadequate evaluation of the processes being discussed.