Adaptive immune system did not evolve by mutations and selection
Dr Alfred Schurmann - computer scientist and mathematician

  See also 1)    Multicellular animals did not evolve from unicellular organisms 
                2)     Human did not evolve by random cell alterations and selection  (added March 23, 2009)
3)     Idea of the anti-Darwin theory of devoloping of life  ; in German, only summary is in English (line added June 15, 2009)

1. Introduction. It is uncertain how the adaptive immune system evolved (s. J. Neumann [NEU]: Immunologie, 2008,Kap. 7.6). At the first sight, it evolved evolutionary from an ancestor system (as suggested in B. Lewin [LEW]: Genesis IX, 2008, Sec. 23.20, in J. Klein [KLE]: Der Stammbaum der Wirbeltiere und der Ursprung des adaptiven Immunsystems (in German), and in T.A. Brown [BRO]: Genomes3, Garland Science Publishing (2007), Sec. 18), because: (a) non-vertebrates have innete (nonadaptive) immune system, (b) some genes (i.e. mete-programs), e.g.  SpRag1L and SpRag2L, which occur in sea urchin are homologous to genes (meta-programs), e.g. Rag1/2, used  in adaptive immune system (s. J. Neuman [NEU], Kap. 7.6), (c) some jawless fishes (lamprey and hagfishes) have cells that resemble lymphocytes of jawed vertebrates very closely (s. J. Klein [KLE]), but they are not capable of adaptive immune response, (d) jaw vertebrates have adaptive immune system. However, if we think over this reasoning, then we may have serious doubts whether it is correct. Firstly, we have no evidence that random mutations and selection changed some genes in non-vertebrates into genes of adaptive immune system. Secondly, such reasoning does not solve the problem of the origin of adaptive immune system, because we do not know what properties had this ancestor system and how it evolved. Also, it is only a belief that an ancestor transposon (as suggested in B. Lewin [LEW], Sect. 23.10) could transform genes into genes (meta-programs) for recombination of immune genes. As J. Neumann [NEU] (Sec. 7.6) pointed out, the question of the origin of adaptive immune system has not been answered.
         Below, I prove that actually, adaptive immunity did not evolve evolutionary by random mutations and selection. First I characterize some essential properties of adaptive immune system. Then I list the possible mutations (operations), which may alter an animal system. Then I show that a finite sequence of random mutation operations cannot transform meta-programs (genes) which do not belong to adaptive immune system in very complex and integrated meta-programs belonging to adaptive immunity.

2. Some essential properties of adaptive immunity

       Let AIP denotes family of meta-programs (i.e. eukaryote genes) having the following properties:
Ip1.    The adaptive immune system contains lymphocyte cells B or T and the major histocompatibility locus (called MHC locus); each of these components contain many linked meta-programs belonging to the family AIP, which produce many simpler meta-programs of the adaptive immune system. B lymphocytes have meta-programs (i.e. eukaryote genes) which assemble simpler meta-programs (genes) from DNA segments (which are genes) and these simpler meta-prpgrams produce immunoglobuline (in short Ig) proteins, called antibodies, specific for a foreign molecule (which are associated with antigens); Ig proteins are operations which recognize these antigens and help to remove or kill it. T lymphocytes produce receptors (denoted by TCR) which function as antibodies produced by B-cells. MHC-loci meta-programs produce Mhc proteins (operations); a Mhc operation presents antigen to T- cell receptors and in this way TCR recognizes this antigen. Each B-cell and T-cell activates only one meta-program which produce a single Ig proteine or a single TCR receptor, respectively.
Ip2.     There are meta-programs in AIP which assemble:
i.    metaprograms V-J-C (called lambda or kappa light-chain genes) from V gene segments (i.e. from program modules)  and J-C gene segments; meta-program V-J-C (a gene) comprises a V gene segment, a J gene segment and a C gene segment,
ii.    meta-programs V-D-J-C (called heavy-chain genes) from V gene segments, D gene segments, J gene segments and C gene segments.
Meta-programs V-J-C and V-D-J-C are executed and produce lambda or kappa light-chain proteins (operations) and heavy-chain proteins (operations), respectively.
Ip3.      T-cells have similar meta-programs as B cells.
Ip4.       No meta-program of the AIP family belong to a nonvertebrate organism.
        As said in Introduction, non-vertebrates have not meta-programs of adaptive immunity. Adaptive immune system is described in B. Lewin [LEW], 2008, and D.P. Snustad & M.J. Simmons [SMS], 2006.

3. Mutations of non-vertebrates
        Mutations are operations which transform DNAs and chromosomes; in this way mutations (operations) may transform an organism system. Below, I list mutations which may change a non-vertebrate system (s. P. Snustad & M. Simmons [SMS], 2006, and A. Schurmann [ASC], 2008;
Postreplic-rep-mut: when a repair scan/procedure detects a damaged DNA (e.g. a thymine dimer in DNA strand) then
it interrupts the process and restarts the synthesis by using a homologous DNA strand; the part of the original nucleotide sequence beginning at the gap is lost (s. [SMS]).
SOS-response-mut: it is a family of procedures (called the error-prone-repair system) which are applied when DNAs are heavily damaged; SOS-genes (programs) are activated - they repair and replicate damaged DNAs using related nucleotide sequences; these replications are not accurate, so replication errors occur; thus the SOS-response build gene mutations (s. [SMS]).
Exis-rep-mut : this procedure excises the damaged bases in DNA; then it fills in the gap by using the undamaged complementary strand of the DNA; and then it seals the break.
Inversion: a chromosome segment is detached, flipped around 180, and reattached to the rest of the chromosome. It can be induced by X-irradiation, transposable elements or by mechanical agents.
Recombination: a DNA segment, z, is moved to an other DNA (and chromosome) in place of a segment, y, and simultaneously segment y is moved to the place where segment z has been.
Transpos-elem: there are three main types of transposable element mutations in eukaryotes:
a)     insertion-sequences (or IS elements): an IS element is excised from chromosome or mitochondria and then inserted into another position at the same or different DNA, where this element itself controls this process;
b)     composite and Tn3 transposons: they are able to copy and insert genes in other DNAs.
c)     retrotransposon - it uses RNA, associated with this transposable element, as a program and makes-up a sequence of DNA molecules; then it inserts this sequence into a new chromosome site.
Transduction: it is a mutation caused by viruses or bacteria - they destroy a cell (i.e. degrade its DNAs) or transfer non-vertebrate genes into other non-vertebrate genes, or integrate a DNA of a virus or bacterium in a non-vertebrate chromosome.
Nonsense-mut: it is a mutation of a termination triplet or it produces termination triplets within genes; these mutations are often non-functional.
Missense-mut: this mutation changes a triplet in a gene (i.e. an instruction in a program) so that it specifies a different aminoacid, and the mutated gene (altered program) produces altered (at one place) polypeptide.
Suppressor-mut: it is a mutation in tRNAs that suppresses other mutations (e.g. some nonsense mutations).
Induced-mut: they are caused by physical or chemical agents; radioactive radiations damage and degenerate DNAs; some such degeneration are repaired by said repair procedures; a chemical agent can transform non-vertebrate gene only into an other non-vertebrate gene, or into a degenerated or damaged non-functional gene.
Parental-DNA-recombination: mutation caused by breakage and crossing over of homologous parental DNA parts and then rejoining some of them in new combinations.
Germinal mutations. Only mutations of germ-line cells may affect the progeny of an animal. The same molecular change in a gene fragment in a germ-line cell may have the following different effects in the progeny, depending on the function of this gene fragment:
a)      if gene is involved in production of
a protein which is not essential for the metabolic or replication process, then its mutation does not lead to non-functional progeny;
b)      if gene is involved in production of a  protein needed in metabolic, then its mutation (which does not produce the needed protein) blocks the metabolic pathway in the progeny;
c)      mutations in gene segments ivolved in a control or replication procedure (i.e. a meta-program implemented in genes) alter this procedure (i.e. genes) in most cases to a wrong or non-functional one,
d)       mutations in a gene (meta-program) which is switched off (i.e. it is not involved in normal living processes of the progeny) do not change the living processes of the progeny - such mutations can influence the progeny only if this meta-program (gene) is activated (e.g. by the repair system),
e)       mutations of genes involved in a repair procedure (ie. in genes where this procedure is implemented) alter this procedure, in most cases to a wrong or non-functional one,
f)        mutations caused by parental-DNA-recombination recombines some parts of homologous DNAs of two similar non-vertebrate systems in such a way that the resulting system is too a non-vertebrate system or it is a non-functional one.
4. Adaptive immunity did not evolve by random mutations
         The above listed mutations are all operations by which multicellular animal progenies can be transformed. Let us note that before vertebrates arose, there was no meta-program belonging to the AIP family (AIP was empty). Now we may formulate the problem we are considered with as follows:
Problem: could a finite sequence of germinal random mutations transform some meta-programs implemented in genes of non-vertebrates into meta-programs (genes) belonging to the family AIP?
Below I prove that the answer is "no".
        Let NVA denotes a multicellular animal living in time when no animal had a meta-program belonging to AIP.
I.     First, we consider the case when a germinal random mutation of type  (a) or (b) or (c) or (e) or (f) occured. Thus it is a mutation of a gene (meta-program) which is not switched off (i.e. it is in an activated state). By such mutation system NVA is transformed into a system NVA1.
i.      If this mutation is inversion, recombination, transpos-elem type (a) or (b), missence-mut or suppressor-mut, then NVA1 is a non-functional system or it is also a system of type NVA, because these operations cannot transform a meta-program (or a sequence of instructions) belonging to NVA into a meta-program belonging to AIP; but some of its genes (meta-programs) may be altered to type (d) (i.e. they have been switched off) or some of switched off genes may have been switched on (it is they may be activated); the last case is considered below in  (II).
ii.      If the mutation is transduction, then NVA1 is either a destroyed system or it contains only genes or pseudogenes which occur in NVA or bacteria or viruses (in NVA1 some genes may have been switched off); thus in NVA1 is no meta-program which belongs to AIP, so NVA1 is of type NVA. It is possible that some genes have been activated; this case is considered in (II).
iii.      If the mutation is postreplic-rep-mut, exis-rep-mut or SOS-response-mut, then it leads to a  non -functional  or mutated system  in which  some meta-programs (genes) may have been switched on (activated);  as explained in (II), the repair procedures activate only meta-programs which do not belong to AIP.
iv.     If the mutation is a nonsense-mut or induced-mut, then NVA1 is a destroyed system (with degenerated DNAs) or it is in a damaged state in which it activates its repair procedures said in (iii).
v.      If the mutation is a retrotransposon, then NVA1 is also a NVA like system (we assume that in NVA is not a program for developing a meta-program belonging to AIP). Some genes may have been activated; this case is considered in (II).
vi.     If the mutation is parental-DNA-recombination, then  the system NVA1 contains only meta-programs consisting of recombined parts of two similar DNA systems; such meta-programs do not belong to AIP; thus NVA1 is of type NVA.
II.      Now we concider the case when random mutation is of type (d), i.e. switched off sequence of instructions (genes or pseudogenes) have been mutated and the mutated NVA system is functioning. Switched off genes and pseudogenes (i.e. sequences of instructions) mutade randomly and there is no selection as long as these sequences of instructions are in the switched off state. Sequence of such mutations do not convert to any functioning program; random mutations transform a sequence of instructions (a gene or pseudogene) into non-functional sequence of instructions or non-interpretable molecule sequence; although, by probability theory, there exist randomly obtained instruction sequences which are programs; but probability theory does not model mutations of genes exactly. In our case, one can only believe that, more than 500  million years ago, sequences of germinal random mutations of genes or pseudogenes (in inactivated genes, where selection did not work) led to more than 50 very complex and integrated meta-programs belonging to the family AIP. The fact that germinal mutations did not lead to meta-programs of adaptive immunity in non-vertebrates shows that the said belief is not scientific grounded.
Thus, the switched on genes or pseudogenes, derived by above said mutations, do not store meta-programs from AIP. Thus, the adaptive immune system did not evolve from accumulated germinal mutations.

[BRO]    T.A. Brown : Genomes3, Garland Science Publishing, (2007).
[KLE]     J. Klein : Der Stammbaum der Wirbeltiere und der Ursprung des adaptiven Immunsystems (in German), Bericht 2003, Max-Plank-Institut fuer Biologie; Tuebingen, Germany 2003.
[LEW]   B. Lewin : Genes IX; Jones& Barklert Publ., USA; 2008.

[NEU]     J. Neumann: Immunbiologie; Springer Verlag Berlin Heidelberg; Germany; 2008.
(SCH]      A. Schurmann: Multicellular animals did not evolve from unicellular organisms; , 2008
[SMS]    D. P. Snustad & M. J. Simmons; Principles of Genetics; John Wiley & Sons, Inc., USA (2006).

To the root site  

Alfred Schurmann                                                  
Copyright  January 23, 2009;    corrected January 27, 2009