The Medical Implications of Complete Mitochondrial DNA Sequencing

 

Ian Logan

 

 

The mitochondrial DNA sequencing for the genealogical community until very recently was only of parts of the genome considered to be medically unimportant; however anyone can now have a test that will provide the complete mitochondrial DNA (mtDNA) sequence.  But the results of this test may have medical implications; and in this article the factors showing this are discussed. The subjects of mitochondrial science, mutations, mutation lists, protein coding and transfer RNA are introduced to provide a background to understanding the origin of various mitochondrial diseases.  The condition of Leber's Hereditary Optic Neuropathy and the diseases associated with mutations in the tRNA for the amino acid, isoleucine, are used as examples.  Two complete mtDNA sequences are analysed to show how a report on their mutations can be produced; and the risk to a subject of getting a report listing potentially harmful mutations is discussed. Mention is made of various instances when complete mitochondrial sequencing may be requested under the guise of genealogy when it really is being done for medical reasons.  Complete mitochondrial DNA sequencing should not be undertaken without careful consideration.

 

 

Received:  October 2, 2005, accepted November 2, 2005

 

Address for correspondence:   Ian Logan, [email protected]

 

 



Introduction

 

The sequencing of mitochondrial DNA (mtDNA) has become very popular with both amateur and professional genealogists over the last few years.  This subject has been made particularly interesting by Professor Bryan Sykes's book, The Seven Daughters of Eve (Sykes, 2001) in which he discusses the concept that all people of European descent have as their female ancestors just a few women who lived many thousands of years ago.  His ideas are part of the general theory that all of mankind is descended from a “Mitochondrial Eve” who lived in Africa about 180,000 years ago.

 

However, the mtDNA sequencing that has been done until recently was only of parts of the genome considered to be medically unimportant; by which is meant that there was no proven connection between the results obtained and known illnesses.  Therefore, sequencing of mtDNA was considered as harmless to the person being tested and there was no reason for anyone to have any hesitation over being tested.  However, it is now possible to sequence the whole of the mtDNA genome, rather than just 4-8%; and the medical implications are beginning to give concern.

 

Many medical conditions are already known to be caused by changes in the mtDNA and the whole area of knowledge continues to expand rapidly.  So now it is perhaps not a question of having abnormal mtDNA as opposed to normal, but rather a matter of what harm is being caused to a person's health by the particular mtDNA they have inherited.

 

It is therefore no longer possible to say that sequencing the mtDNA genome is a harmless procedure; and in this article the factors that show this are discussed. However, because the underlying science may be new to many readers who do not have a background in a science, there are a number of basic scientific details that have to be explained.

 

 

2. Mitochondrial Science

 

Each mature nucleated cell of the body contains several hundreds of  mitochondria (Taylor 2005).  These are small structures involved in energy production within the cell.  A mitochondrion may be considered as a sort of power plant, with fuel being taken in and energy produced. In each of the mitochondria there are circular strands of genetic material made up of deoxyribonucleic acid.  This is the mitochondrial DNA, or mtDNA.  The structure of one of these strands was first determined in 1981 at the University of Cambridge (Anderson 1981). This structure can be called the mtDNA genome, or the mtDNA sequence.  Figure 1 shows a simplified view of mtDNA.

 

 

 

Figure 1  A Simplified View of mtDNA.

(Adapted from: http://www.mitomap.org/)  The circular strand of mtDNA is shown with nucleotide base 1 at the 12 o'clock position. By convention, the numbering of the nucleotide bases goes in an anti-clockwise direction.  The coding areas for the genes are shown in red, with their corresponding labels outside (see Table 6 for the full names).  The tRNA's are shown in blue with the three-letter abbreviations for the amino acids inside (see Table 7 for the full names).

 

 

 

The original subject was a person of European descent and her sequence now forms the Cambridge Reference Sequence (CRS).  In this sequence the nucleotide bases are numbered from 1 to 16,569; and subsequently, all other mtDNA sequences have been compared to the CRS.   Differences between sequences are normally given as a simple list.

 

 The purpose of the mtDNA is to carry encoded instructions for certain products in the cell.  In it, there are coding regions for two ribosomal RNA genes, 13 genes for the production of proteins involved in oxidative phosphorylation of fatty acids and carbohydrates (Taylor 2005), and, perhaps most interestingly in the present context, the coding for 22 transfer RNA (tRNA) genes for the 20 amino acids that are the building blocks of proteins.  There are also parts of the mtDNA that appear to have no function, other than perhaps to ensure the correct shape to the overall circular form of the mtDNA. The largest of these non-functional areas contains the hypervariable regions, HVR-1 and HVR-2.

 

The actions of mitochondria are mainly concerned with the energy processes within cells; and whilst a single mitochondrion is very small and has an insignificant influence on any particular cell, the cumulative effect of mitochondria on health and illness in a person can become clear when one considers that about 2 billion mitochondria are made every second throughout a person’s life.  This estimate is obtained by considering the human body as having 75 trillion cells, each with 250 mitochondria and having an average life of around 100 days.

 

Until recently sequencing of the mtDNA for genealogical purposes was restricted to the hypervariable regions; and many thousands of subjects have already been tested.  But now it is possible to sequence the whole mitochondrial genome for any individual at a reasonable cost; in addition, already about 2,500 complete mtDNA sequences have been published from scientific studies.  It is the data from these published complete mtDNA sequences that provide the material for this article.

         

 

3. Mitochondrial Nucleotides

 

The mtDNA genome is considered as a circular strand of DNA consisting of 16,569 nucleotide bases and for historical reasons the numbering of the bases begins at an arbitrary place—the start of what is termed “the hypervariable HVR-2 region.”

 

As in other DNA, there are 4 nucleotide bases, Adenine, Guanine, Cytosine and Thymine; and from now on these will be referred to by their initial letters, i.e. A, G, C and T.  The actual mtDNA is made up of a double strand, but for simplicity only one strand is normally shown.  The second strand is complementary by which is meant that a 'T' links with an 'A', an 'A' with a 'T', a 'C' with a 'G' and a 'G' with a 'C'. The pairings of A-T and C-G are known as “Watson-Crick base pairings” after the discoverers of the structure of DNA.

 

It is a feature of mtDNA that some of the coding is read on one strand going forward in one direction, whilst other coding is read from the complementary strand in the opposite direction.

 

It is not practical to print the full sequence of 16,569 bases that makes up the Cambridge Reference Sequence (CRS) in this article, but Table 1 is included to show the first 70 nucleotide bases.  The entire sequence may be found at:

 

http://www.mitomap.org/mitoseq.htm

 

 

4.  Mitochondrial Mutations

 

mtDNA sequences from persons who are maternally related to each other are normally identical; but it is found that once in about every 2,000 years in a given lineage, one of the 16569 nucleotide bases alters.  This is a random process.  Such an alteration is termed a mutation.

 

Table 1

First 70 Bases of the Cambridge Reference Sequence.

 

The bases are numbered from 1 to 70 and form the first part of the hypervariable region HVR-2. The sequence is often split into groups of 10 bases to aid counting.

 

gatcacaggt   ctatcaccct   attaaccact  cacgggagct  ctccatgcat  ttggtatttt  cgtctggggg

1            11           21          31          41          51          61

 


In most instances the change is of a “C” being replaced by a “T”, or vice-versa, or a “A” being replaced by a “G”, or vice-versa, and uncommonly the remaining possibilities.  It is also possible for nucleotide bases to be added—these are called insertions, or for bases to be lost—these are called deletions.

 

Mutations cause the mtDNA sequences from persons who are not closely related to be different.  And, it follows that the number and type of mutations between two persons gives a measure of their relatedness. Consequently, two persons who are both European, for example, will have fewer mutational differences as when compared with persons who are from very different parts of the world.

 

To illustrate this point, Table 2 shows a typical list of differences from the CRS.  The results come from an Italian (Achilli 2004).  The list here is short as the subject is being compared against the CRS individual who is considered to come from a more northerly part of the same continent, Europe.  The length of the list suggests a common matrilineal ancestor perhaps 20-30,000 years ago.

Table 2

A Typical "Mutation" List for a European

 

The mtDNA from an Italian subject shows differences from the CRS at the following places:

 

   T152C   A263G   315.C   A750G  A1438G  G3591A  A4310G  A4769G  A8860G

  T9148C T13020C A15326G C16168T T16519C

 

Note: The T152C shows that the 'T' at position 152 in the CRS is replaced by 'C' and the 315.C shows there is an insertion of a 'C' between 315 & 316.

 


 

Table 3 shows the list of differences from CRS for a person from Uganda (Macaulay 2005).  In this case the mutation list is very much longer; and supports the theory that the earliest African lineage, Haplogroup L0, and Europeans, as typified by the CRS, have been separate for approximately 180,000 years.  Kivisild (2005) gives the figure of 160,000 (range 138,000 - 182,000) years, based on their study of 277 complete mtDNA sequences.  But in the author's view the higher figure is to be preferred, and is supported by other studies (Macauley 2005).

 

 

 

Table 3

A More Complicated "Mutation" List

 

The mtDNA from an Ugandan subject, from Haplogroup L0, shows differences from the CRS at the following places:

 

     G143A    T146C    T152C     G185A     A189G     G247A     A263G     315.C     A750G

     G769A     825A    A978G    G1018A    C1048T    A1438G    G1719A    A2245G    A2706G

    G2758A   C2772T   C2789T    T2885C    C3107T    C3516A    C3594T    C3852T    A4104G

    C4194T   C4312T   A4562G    T4586C    A4769G    C4964T    C5321T    T5442C    C5603T

    T6185C   A6359G   C7028T    A7146G    T7148C    C7256T    G7521A    C8468T    C8655T

    A8701G   A8860G   C9042T    A9347G    T9540C    T9581C    C9620T    C9818T   G10143A

   A10398G  G10589A  C10664T   G10688A   T10790C   T10810C   T10873C   T10915C   T11287C

   T11299C  A11641G  G11719A   G11914A   G12007A   C12705T   A13105G   A13276G   A13470G

   C13506T  C13650T  C13680T   G13708A   G13928C   C14109T   C14620T   C14766T   C15136T

   A15326G  G15431A  T15852C   G16129A   C16169T   T16172C   C16173T   C16187T   T16189C

   C16223T  A16230G   16239T   C16278T   T16311C   C16327T   T16368C   T16519C

 

 

 

 

5. Phylogenetic trees

 

By studying the mutations that are found in different sequences it is possible to draw a tree to show how different populations have branched away from the line of mutations that leads from 'mitochondrial Eve' to the European person whose sequence has been chosen as the CRS.

 

Each of these populations has been given a label called a haplogroup; and the CRS sequence belongs to Haplogroup H2, the Italian sequence, above, to Haplogroup H9 and the African sequence, above, to Haplogroup L0.

 

Table 4 shows the probable mutations from “mitochondrial Eve” (mtEve) to the CRS. All the mutation sites are to be found in the Ugandan’s mutation list, given above, as Haplogroup L0 is the first population that branches off.  However, the Ugandan subject’s list of differences from CRS has twice as many mutations as appear in Table 4 because his list must include not only his mutations from mtEve, but also the reverse of the mutations between mtEve and CRS.

 

 

Table 4

Mutation Sites from “Mitochondrial Eve” to the CRS

 

(Omitting hypervariable areas)

 

Line from Branch

Point to Branch Point*        Mutations Along this Line at these Locations

 

From mtEve to BP-to-L1        4312   6185   9755  10589  11914  12007

From BP-to-L1 to BP-to-L5     2758   2885   7146   8468

From BP-to-L5 to BP-to-L2      825   8655  10688  10810  13105   13506   15301

From BP-to-L2 to BP-to-L4     3594   4104   7256   7521  13650

From BP-to-L4 to L3            769   1018    

From L3 to N                  8701   9540  10398  10873  15301

From N to R                  12705

From R to HV                 11719  14766

From HV to CRS                 750   1438   2706   4769   7028   8860  15326 

 

*  “Branch Point” or “BP” means a point on the line in the mtDNA phylogenetic tree between mtEve and CRS, where a branch leads to another named haplogroup.  These branch points are shown in Figure 1 in the article by Kivisild (2005).  A detailed discussion of the reasons for including each of the mutations given above will be the subject of a forthcoming article in JoGG.

 

Note: 15301 has changed twice.

 

 

 

It is important, however, to appreciate that if the sequence chosen as the CRS  had been the African sequence, as given above, European sequences would be seen as very different to the reference sequence  - it is just a matter of history, and with hindsight perhaps the use of the hypothetical reference sequence for 'mitochondrial Eve' would have been best from all points of view. However, in the remainder of this article, the normal convention—representing a sequence as a set of differences from CRS—will be followed. Using this convention, the word "mutation" will mean difference from CRS.

 

 

6. The Genes Encoded in the mtDNA

 

The mtDNA genome has 15 areas of coding for genes. There are two genes which are involved in ribosomal function (ribosomes are other small structures in a cell) and 13 genes concerned with the the biochemical process of oxidative phosphorylation.

 

The two ribosomal genes code for ribosomal RNA. In this article the function of the genes will not be discussed further as there does not appear to be any definite medical condition associated with any of the many known mutations in these areas of coding.  Presumably these mutations do not affect overall  ribosomal function to any significant degree.  The mutations are, however, useful when constructing phylogenetic trees.  Table 5 gives the details of these genes.

 

 

Table 5  The Names and Locations of the Ribosomal Genes Found in mtDNA.

Adapted from: http://www.mitomap.org/

 

Name:                                    Location

 

12S ribosomal RNA                        648 – 1601

16S ribosomal RNA                       1671 – 3229

 


 

 

However, the 13 genes that encode for proteins involved in oxidative phosphorylation are very important as they are involved with many medical conditions. These genes are listed in Table 6.

 

Table 6

The Names and Locations of the Oxidative Phoshorylation Enzyme Genes Found in mtDNA

(Adapted from: http://www.mitomap.org/)

 

Name:                                             Location:

 

NADH dehydrogenase subunit 1 (ND1)         3307  -  4263

NADH dehydrogenase subunit 2 (ND2)         4470  -  5513

NADH dehydrogenase subunit 3 (ND3)        10059  - 10404

NADH dehydrogenase subunit 4L (ND4L)      10470  - 10766

NADH dehydrogenase subunit 4 (ND4)        10760  - 12137

NADH dehydrogenase subunit 5 (ND5)        12337  - 14148

NADH dehydrogenase subunit 6 (ND6)        14149  - 14673  (Complement)

cytochrome c oxidase subunit I (COX1)      5904  -  7445

cytochrome c oxidase subunit II (COX2)     7586  -  8269

cytochrome c oxidase III (COX3)            9207  -  9987

ATP synthase F0 subunit 6 (ATP6)           8527  -  9207

ATP synthase F0 subunit 8 (ATP8)           8366  -  8572

cytochrome b (CYTB)                       14747  - 15886

 

Note:  Because of post-transcription processing of the last few bases in the RNA copy of the DNA, the number of locations is not always a multiple of three (see also Note 2 to Table 7).

 


 

 

These genes all code for protein using a triplet-codon system, which means that a group of 3 consecutive nucleotides identify an amino acid for the protein.

 

To illustrate this, 18 nucleotide bases from near the start of the gene ND1 (beginning at 3331) are shown with the corresponding amino acid coded by each triplet of bases:

 

 

CTC   ATT   GTA   CCC   ATT   CTA   ….

 

Leu   Ile    Val    Pro    Ile    Leu   ….

 

 

Table 7 shows the full details of the triplet-codon system.

 

Table 7

The Triplet-Codon System Used in Genes to Code for Amino Acids

 

Amino Acid           Triplet-codons     Amino Acid         Triplet-codons

 

Alanine(Ala)         GC*                Leucine(Leu)       TTA   TTG   CT*

Arginine(Arg)        AGA   AGG   CG*    Lysine(Lys)        AAA   AAG

Asparagine(Asn)      AAC   AAT          Methionine(Met)    ATA   ATG   ATT

Aspartic Acid(Asp)   GAC   GAT          Phenylalanine(Phe) TTC   TTT

Cysteine(Cys)        TGC   TGT          Proline(Pro)       CC*

Glutamic acid(Glu)   GAA   GAG          Serine(Ser)        AGC   AGT   TC*

Glutamine(Gln)       CAA   CAG          Threonine(Thr)     AC*

Glycine(Gly)         GG*                Tryptophan(Trp)    TGG

Histidine(His)       CAC   CAT          Tyrosine(Tyr)      TAC   TAT

Isoleucine(Ile)      ATC                Valine(Val)        GT*

 

And the beginning and end of coding areas usually have one of the triplet-codons:

 

START                ATA   ATG   ATT    STOP               TAA   TAG   TGA

 

Note (1):  An asterisk (*) indicates any of A, C, G, or T.

Note (2): For simplicity, 'T' is used here as this is the nucleotide in mtDNA, although a uracil (U) is used in the RNA that is transcribed.  Also note that mitochondria use a slightly different genetic code from nuclear DNA—ATA and ATT code for methionine and START instead of the normal isoleucine.  The STOP codons are sometimes produced only in post-transcriptional processing, so some mtDNA genes will not have a normal STOP code at the end location of the gene.

 


 

 

But what is of interest here is the way in which mutations can affect the coding. A mutation may lead to the gene encoding a different amino acid at a particular place, with the potential for impairing the protein's function. However, it is also possible that the effect of a mutation can be neutral, as in the case where a mutated codon codes for the same amino acid.

 

To show this here are three examples of mutations occurring in the coding areas for genes:

 

First, a mutation that does effect a change:

 

The mutation A15326G alters the 1st nucleotide base of the triplet-codon for the 193rd. amino acid of the gene CYT-B.  At locations 15326-15328 in the CRS is ACA, which is a triplet code for threonine, and the mutation A15326G changes the triplet code to GCA and the amino acid to alanine.  This is usually shown as A15326G …. Thr > Ala to indicate the amino acid changes.

 

Secondly, a neutral mutation:

 

G15301A alters the third nucleotide base of  the triplet-codon for the 184th amino acid of the gene for CYT-B.  Locations 15299-301 in the CRS has TTG, a triplet-codon for leucine, and the mutation G15301A changes this to TTT, which is also a triplet-codon for the amino acid leucine.  This can be written as G15301 .. Leu >Leu to indicate the amino acid does not change.

 

And thirdly, a mutation of medical interest:

 

The mutation T14484C is associated with a form of blindness (discussed below) and has an occurrence of no more than 0.01%.  T14484 alters the first base of the triplet code for the 64th amino acid of the protein gene for ND6.  But note this gene is read from the complementary strand.

 

Locations 14484, 14483 & 14482 contain TAC, which when complemented on the other strand becomes ATG, a triplet-codon for methionine.  The mutation T14484C changes the triplet-codon to CAC, and the complement of this into GTG, which is a triplet code for valine.  This can be written as  T14484C …. Met > Val.

 

More examples of the effects of mutations in the coding areas for the genes will be given later.

 

 

7. The transfer RNA coding areas

 

In a cell a protein is assembled from its constituent amino acids according to the order specified by the gene for that protein. It is therefore necessary for the cell to have an adequate supply of each of the different amino acids available to be used. However, there also needs to be a method of handling the amino acid molecules; and the system that has evolved to do this involves the use of small pieces of ribonucleic acid (RNA). There is a slightly differently coded piece of RNA for each amino acid and these pieces of RNA are called the transfer RNAs (tRNA).

 

The structure of each of the transfer RNAs is very similar one to another and when the pieces are free within the cell they all take up a cloverleaf shape. The stem and side leaves of a transfer RNA binds to its appropriate amino acid whilst the end of the middle leaf is termed the anti-codon, as it is this area that binds with a gene's triplet-codon. For example, the anti-codon for isoleucine is GAT, which once complemented and reversed becomes ATC, which is a triplet-codon for isoleucine.

 

Table 8 gives three examples of the sequences that code for tRNAs. Each is aligned to show the structural similarities. The anti-codon areas are shown in bold and underlined.  For interest a sequence from another mammalian species is also shown.  In the mtDNA there are 22 areas that code for the tRNAs; and the details are given in Table 9.

 

 

Table 8

tRNA Sequences From the CRS

 

Alanine: (complemented and reversed from CRS)

AAGGGCT TA GCTT AATTA  AAGT G GCTGA TTTGCGT TCAGT TGAT  GCAGA GTGGGGT TTTGC AGTCCTT A

 

Valine:

CAGAGTG TA GCTT AACACA AAGC A CCCAA CTTACAC TTAGG AGAT  TTCAA CTTAAC  TTGAC CGCTCTG A

                                              

Isoleucine:

AGAAATA TG TCTG ATAA   AAGA G TTACT TTGATAG AGTAA ATAAT AGGAG TTTAAAC CCCCT TATTTCT A

 

And, for interest as it illustrates a point that will be mentioned later, the following is the sequence for the tRNA of isoleucine from a Hill Wallaroo, a species of Kangaroo! (Janke 1997)

 

AGAAATA TG TCTG ACAA   AAGA A TTATC TTGATAG GATAA ATTAT AGGGG TGCAAGC CCCCT TATTTCT A

 

Note (1):  The anti-codon areas are shown in bold and underlined.

Note (2):  The table shows each tRNA sequence grouped into 15 functional parts.

Note (3):  Figure 2 shows the tRNA for isoleucine drawn as a two-dimensional “cloverleaf”

 


 

Table 9

The Locations of the Transfer RNAs 

(Adapted from: http://www.mitomap.org/)

 

  1. Phenylalanine  577 -   647          12. Serine(UCN)    7446 -  7514 (Complement)

  2. Valine        1602 -  1670          13. Aspartic acid  7518 -  7585

  3. Leucine(UUR)  3230 -  3304          14. Lysine         8295 -  8364

  4. Isoleucine    4263 -  4331          15. Glycine        9991 - 10058

  5. Glutamine     4329 -  4400 (Compl.) 16. Arginine      10405 - 10469

  6. Methionine    4402 -  4469          17. Histidine     12138 - 12206

  7. Trytophan     5512 -  5579          18. Serine(AGY)   12207 - 12265

  8. Alanine       5587 -  5655 (Compl.) 19. Leucine(CUN)  12266 - 12336

  9. Asparginine   5657 -  5729 (Compl.) 20. Glutamic acid 14674 - 14742 (Complement)

 10. Cysteine      5761 -  5826 (Compl.) 21. Threonine     15888 - 15953

 11. Tyrosine      5826 -  5891 (Compl.) 22. Proline       15956 – 16023 (Complement)

 

Note:  The two areas of coding for Leucine and Serine match against different anti-codons.

 

Drawn in Figure 2 is the cloverleaf shape for just one tRNA, that for isoleucine, using the sequence from the DNA instead of the corresponding RNA.  Note that, except for just one instance, the stems of the leaf show Watson-Crick pairings, ie., with A-T and C-G pairings.  It is thought that these pairings in the stems add to the stability of the structure.  The sequence coding for this tRNA is at 4263-4331 in the mtDNA.  The three bases at 4292-4294 are termed the anti-codon.  The 'GAT', once complemented and reversed, gives the triplet-codon, 'ATC,' which codes  for isoleucine (as shown in Table 7).

 

 

 

Figure 2  The tRNA for isoleucine drawn in a 2-dimensional 'cloverleaf' shape.

 

 

 

8. Mitochondrial Diseases

         

It is not intended here to discuss mitochondrial diseases in great detail, but there are several points to be made initially.

 

Firstly, the term itself can be used to include both the diseases that arise from the malfunctions in the mitochondria that are the result of errors in the nuclear chromosomal inheritance and also for diseases that are caused by errors in mtDNA inheritance.  In this article only diseases that result from changes in the mtDNA genome itself will be considered.

 

Secondly, there is the fascinating problem of homoplasmy and heteroplasmy - by which is meant that a particular subject with all of their mtDNA strands identical one to another is termed homoplasmic; whereas, a person is said to be heteroplasmic if they have both a population of mtDNA strands that are one type and also a population of strands showing a mutation.

 

And thirdly, there are also very rare occasions when a severe illness can arise from major rearrangements, or deletions, of the mtDNA, but these will not be discussed further.

 

Disease associated with coding for genes:

 

Quite a large number of conditions are associated with mutations in the coding for genes, but for the most part each condition is very uncommon. However, the condition of Leber's Hereditary Optic Neuropathy (LHON) is common enough to warrant discussion. LHON is a well recognised condition and there are about 2,000 cases in  the UK (among a population of about 60 million).

 

LHON is a condition that causes severely impaired vision and it was first described by Theodore Leber in about 1870.  He, of course, had  no understanding of the mode of inheritance and, indeed, this was still unknown until the 1980's.  Now the condition is known to be associated with a number of mutations in the mtDNA.

 

In this condition the typical presentation is for an otherwise fit young man to lose a large part of his vision over just a few months; and once lost the vision does not recover.  But although most common in young men, the condition can affect persons of all ages, both sexes and of all races.  And, like with so many other mitochondrial diseases the condition can be mild or severe, fluctuate in the degree of its severity with time, and appear to skip generations.

 

It is thought the optic nerves (the nerves from the brain to the eyes) are affected in this condition because they consume large amounts of energy.  And, the very slight impairment of function in the production of energy in the mitochondria of the cells in the optic nerves becomes significant.  Some other problems, such as tremor, have been reported in sufferers of LHON, but it is more usual for a person with LHON to be normal except for the severe visual impairment.

 

Some mutations that have been shown to be associated with LHON are:

 

3460A  (ND1), 10663C  (ND4L), 11778A  (ND4), or 14484C (ND6), among others.

 

These mutations can be found in family pedigrees and can also occur sporadically. The fact that several mutations can be involved with the condition suggests that the disease results from a general disruption to the oxidative phosphorylation process rather than from something specific to a particular mutation.

 

Disease associated with the coding for the tRNA's:

 

There are many diseases associated with mutations in the 22 tRNA coding areas and new discoveries continue to be made on a regular basis.

 

But whereas, most of the diseases caused by mutations in the gene coding area are uncommon, the diseases associated with mutations in the tRNA coding areas are very well known.  For example, there are associations between mtDNA mutations and some forms of deafness, eye disease, heart disease, myopathy, strokes and hypertension.  And, quite probably, mtDNA mutations may also be involved in conditions such as fatigue and migraine.

 

Again however, the point has to be made that the mutated mtDNA itself does not cause the disease in the way that cystic fibrosis is caused by a faulty protein because of a mutation in the DNA of a chromosome, but rather the mutated mtDNA may lead to a reduction in the intracellular availability of a particular amino acid and it is this biochemical imbalance in the cells of a particular tissue that results in illness.

 

It is beyond the scope of this article to go through all the conditions associated with the different tRNA's, but it will be helpful to look at a representative tRNA to see what diseases are associated with the mutations in the coding area for that tRNA.  The following discussion takes as its example the tRNA for isoleucine.

 

As shown above the tRNA coding for the amino acid isoleucine is located in the mtDNA from 4263 to 4331.  In this area there are mutations which are pathological (harmful), some that are possibly neutral in their effects, and one that on theoretical grounds could actually be beneficial.  For reasons that are not known it would appear that mutations in the mtDNA coding for the tRNA for isoleucine mainly cause trouble in the heart and the eyes.  Table 10  gives a list of mutations and associated diseases for the tRNA coding area for isoleucine (taken from the Mitomap database (October 2005).

 

 

 

Table 10

A List of Mutations and Associated Disease for the tRNA Coding Area for Isoleucine

 

A4267G      Myopathy                                 (Muscle)

A4269G      Infantile Cardiomyopathy                 (Heart)

T4274C      Ophthalmoplegia                          (Eye)

G4284A      Varied familial presentation             (Various)

T4285C      Ophthalmoplegia                          (Eye)

T4290C      Progressive Encephalopathy               (Brain)

T4291C      Hypomagnesaemic Metabolic Syndrome       (Blood pressure)

A4295G      Cardiomyopathy                           (Heart)

G4298A      Ophthalomplegia                          (Eye)

A4300G      Cardiomyopathy                           (Heart)

G4309A      Ophthalmoplegia                          (Eye)

A4317G      Infantile cardiomyopathy                 (Heart)

C4320T      Encephalocardiomyopathy                  (Brain & Heart)

 


It is appropriate here to discuss the mutations the isoleucine tRNA found in the published sequences of this tRNA and discuss why these mutations might, or might not, be pathological.  The mutations that have been found in the sequences are:

 

A4295G, A4310G, C4312T, A4315T and the insertion 4315.A

 

A4295G is considered to be pathological as seen in Table 10 because it is very close to the anti-codon loop and presumably disturbs its functioning.

 

A4310G can be presumed to be of low importance as it re-establishes a C-G pairing.  The form 4310G is considered to be the older form  and can be seen to be present in the wallaroo sequence, shown above.

 

C4312T can also be presumed to be of low importance as 4312T is thought to be the older form, and 98% of mankind in effect show the mutation 4312C.  However, A4315T and 4315.A, unfortunately, must both be considered to be of possible harmful significance as each alters the shape of a loop of the tRNA.

 

 

9. Analysis of Selected Complete mtDNA Sequences

 

The above discussion in parts 1-8 has been rather detailed but the reader should now have some idea as to how a "mutation" list obtained from a given sequence can be interpreted with each mutation being rated as to the possibility of its causing harm.

 

Note the importance of a mutation will be low whenever its frequency in the world is high, or the effect on the tRNA, or gene, is considered unimportant—either because it stabilises the structure, is a reversion back to a common mammalian form, or does not change the amino acid.  To show this, possible medical consequence reports for two sequences are shown in Tables 11 and 12.

 

 

 

Table 11

First Report

 

Origin: Italy             Haplogroup: H9         Identity: Not revealed

 

Mutation       Genome Area    Frequency:    Effect:      Importance:

List:

 

T152C             HVR-2           High        Nil           Very low

A263G             HVR-2           High        Nil           Very low

315.C             HVR-2           High        Nil           Very low

A750G             12S rRNA        High        ?             Low

A1438G            12S rRNA        High        ?             Low

G3591A            ND1             Very low    Leu > Leu     Low

A4310G            tRNA-Ile        Very low    Form C-Gpair  Low

A4769G            ND2             High        Met > Met     Low

A8860G            ATP6            High        Thr > Ala     Low

T9148C            ATP6            Low         Leu > Leu     Low

T13020C           ND5             Low         Gly > Gly     Low

A15326G           CYT-B           High        Thr > Ala     Low

C16168T           HVR-1           Low         Nil           Very low

T16519C           HVR-1           Common      Nil           Very low

 


 

 

For the present it is unlikely that genealogical testing companies will wish to produce such reports, but the reports here are produced to show how such reports might appear.

 

The first sequence is from a person in haplogroup H9 (Achilli 2004) and has the mutations shown in Table 11. The sequence includes the mutation at A4310G, - which above was shown to be of low importance.  All of the other mutations are similarly of low, or very low importance, supporting the overall conclusion of the report is that none of the mutations suggest a significant medical problem.

 

However, compare this with a second report from a person in Haplogroup K1 (Coble 2004) and has the mutations shown in Table 12.  This sequence includes the mutation A4295G, which above was shown to be associated with cardiomyopathy.  There are also three other mutations that may possibly be important.  The overall conclusion from this report is that the mutations suggest a significant medical risk to the subject.

 

Table 12

Second Report

 

Origin: America                  Haplogroup: K1                  Identity: Not revealed

 

Mutation      Genome Area      Frequency:    Effect:      Importance:

List:

 

A73G              HVR-2             High        Nil         Very low

A263G             HVR-2             High        Nil         Very low

315.C             HVR-2             High        Nil         Very low

C497T             HVR-2             Common      Nil         Very low

523.C             HVR-2             Common      Nil         Very low

523.A             HVR-2             Common      Nil         Very low

A750G             12S rRNA          High        ?           Low

T1189C            12S rRNA          Common      ?           Low

A1438G            12S rRNA          High        ?           Low

A1811G            12S rRNA          Common      ?           Low

A2706G            12S rRNA          Common      ?           Low

A3480G            ND1               Common      Lys > Lys   Low

A4295G            tRNA-Ile          Very Low    Critical    Significant

A4769G            ND2               High        Met > Met   Low

A5711G            tRNA-Asn          Low         Adds C-G    Possible

G6260A            COX1              Low         Glu > Glu   Low

C7028T            COX1              Common      Ala > Ala   Low

A8860G            ATP6              High        Thr > Ala   Low

G9055A            ATP6              Common      Ala > Thr   Low

T9698C            COX3              Common      Leu > Leu   Low

A10398G           ND3               Common      Thr > Ala   Low

A10550G           ND4               Common      Met > Met   Low

G10586A           ND4               Low         Ser > Ser   Low

T11299C           ND4               Common      Thr > Thr   Low

A11467G           ND4               Common      Leu > Leu   Low

T11485C           ND4               Low         Gly > Gly   Low

G11719A           ND4               Common      Gly > Gly   Low

C11840T           ND4               Low         Leu > Leu   Low

A12308G           tRNA-Leu          Common      Harmless    Low

G12372A           ND5               Common      Leu > Leu   Low

T13740C           ND5               Low         Ile > Ile   Low

T13886C           ND5               Low         Leu > Pro   Possible

C14167T           ND6               Common      Glu > Glu   Low

T14502C           ND6               Low         Ile > Val   Possible

C14766T           CYT-B             Common      Ile > Thr   Low

T14798C           CYT-B             Common      Phe > Leu   Low

A15326G           CYT-B             High        Thr > Ala   Low

G15884A           CYT-B             Common      Ala > Thr   Low

T16224C           HVR-1             Common      Nil         Very low

T16311C           HVR-1             Common      Nil         Very low

T16519C           HVR-1             Common      Nil         Very low

 


 

 

10. So is There a Risk From Complete mtDNA Sequencing ?

 

Yes, and the risk from having a complete mtDNA sequencing begins when a person first thinks about ordering the test - as for the first time it is possible to have a test performed for genealogical purposes that can actually affect the person's well being.

 

Before agreeing to mtDNA sequencing it is important to consider just how one might be affected by the results. Fortunately, for about 90-95% of people the results will not show any mutations of medical significance.  But this means that about 5-10% of people can expect results that show a significant mutation—for example, as is shown in Table 12 where the subject has the significant mutation at A4295G.  Because of the random nature of mutations, it is impossible to predict in advance who might receive such results.

 

The figure of 5-10% is the present author's estimate from studying the mutations found in all of the 2,500 published complete mtDNA sequences.  It is possible that some of the sequences have been published by the researchers because of the mutations that the sequences contain—this would make the 5-10% figure too high.  But against this, mutations associated with medical conditions continue to be identified—this would raise the figure.  The author suggests the figure of 5-10% when mtDNA is sequenced for genealogical purposes.

 

Besides the person themselves worrying, should they be telling other people that they are going to be tested?  And, if so, who should be told? Close relatives and doctors, perhaps. But probably not at this early stage an employer or health insurance company.  It is difficult to judge how important pre-testing anxiety might become, but for some people, fear of adverse results will quite properly stop them from undergoing the testing.

 

However, if mtDNA sequencing is undertaken it becomes important to determine just how significant the results might be.  The results for the Haplogroup H9 person shown earlier, demonstrate that test results can suggest there are no adverse medical mutations, and in such a case there would not appear to be any reason for concern.  For the great majority of tests the outcome will be similar.

 

But what about a case such as that for the Haplogroup K1 person (shown earlier)?  Here the results suggest that the subject has a mutation linked to a significant heart condition.

 

In such a situation it would seem important to convey things in their proper proportions and give the person the most complete understanding of the situation as is possible.  In most instances it would be sufficient to explain that most mtDNA mutations do not cause the disease with which they are associated, but rather explain why the particular condition might arise.

 

Just at present it would seem that most genealogical testing companies are not involving themselves in analysing their results for any medical implications; and it is uncertain how such service might be provided.  But what is clear is that counselling people who have been been found to have important mutations will not be easy, and perhaps will lead to as many unanswerable questions being posed, as questions for which answers can be given.

 

But it is not just the person who receives test results indicating possible medical problems who is affected—other family members will also have obvious concerns.  A mother who receives adverse test results will certainly worry about the health of her children; and a younger person will have concerns about their siblings and mother.

 

Medically significant mtDNA mutations should not be ignored, in the author's view, and their possible significance should be discussed with the family's medical practitioners.

 

Overall, the discovery of potentially harmful mtDNA mutations may have many unexpected and undesirable consequences. 

 

 

11. Sequencing for reasons other than for genealogy

 

Having discussed sequencing for genealogical purposes it would now seem appropriate to discuss briefly three occasions when a person may choose to have complete mtDNA sequencing because it is now available at reasonable cost.  The test may be the result of the person's own decision, or perhaps it might be performed on the advice of their doctor.  In these cases the person is less likely to be worried about finding a medically significant mutation—as this is what they are seeking.

The first instance is when a person suspects they might have a particular mtDNA mutation because a maternal relative has an identified mitochondrial condition.  In this situtation the person must be very careful to understand that should the test show up the mutation then they may be at some risk of developing the same condition.  But, a negative result is to be viewed with some suspicion as a single test of a buccal (cheek cell) smear (the usual method of sampling for genealogical testing) is not a very satisfactory method of testing for any condition that may be heteroplasmic.

 

The second occasion is when a person with a medical condition may wish to undergo testing to prove the mitochondrial origin for their illness.  A particular example of this might be a person with failing vision in whom Leber's Hereditary Optic Neuropathy (LHON) is a possibility.  It should be remembered that a negative result cannot be assumed to be reliable because of the possibility of heteroplasmy.  However, the commonest mutation for LHON, T14484C, is usually homoplasmic; and if T14484C is shown then considering the rarity of the mutation it will be likely that the condition is present and as a result the cause of their failing vision.  This might also be the case when considering conditions such as cardiomyopathy, a more general myopathy, or ophthalmoplegia (paralysis of one or more muscles that move the eyes).

 

The third reason that some may choose to have such a test is perhaps the most interesting and the most speculative, since it is the availability of the test itself that may lead to it being ordered by people suffering from a number of medical conditions.

 

If one considers that most mitochondrial diseases are very variable in their effects, it is reasonable to say that sufferers of a number of non-specific illnesses may well have a significantly increased chance of having an hitherto unexpected mitochondrial disease.

 

So what conditions might these be?  What is a non-specific illness?

 

Here are to be considered conditions that do not have a well understood basis to their causation.  Most diseases have symptoms and observable physical signs - so there is usually no doubt about a person suffering from a chromosomal disorder, a bacterial or viral infection, the after-effects of an injury, degeneration conditions, a cancer, and auto-immune disorders.  But what about symptoms of fatigue, weakness, headache, loss of appetite, and conditions lacking any definitely observable signs.  In these situations, whilst there is the possibility of a mitochondrial basis, it is likely that there is little to be gained by sequencing the mtDNA.

 

However, there still remains a significant number of people who have the vague symptoms as mentioned above, but who also have fully accepted physical signs.  A person may well have a single nerve paralysis (a so called mononeuritis multiplex), history of a small stroke or encephalitic episode, heart disease with a proven reduction of cardiac output—but normal coronary arteries, or muscle weakness to the extent that a muscle biopsy may be a consideration. In all these cases the similarity to known mitochondrial diseases is clear.

 

So, what is the chance of there being a mitochondrial mutation as a basis to the illness suffered by a person suffering from this sort of condition?  If it is considered that a person in the normal population has a risk of an adverse mutation of 5-10%, then in a selected group the author considers that the risk might be doubled—at which level complete mitochondrial sequencing could become a very sensible option.  It is, after all, with the reduction in the cost of testing, cheaper to have complete mtDNA sequencing than to have a set of hospital-based tests.

 

Hitherto, the study of diseases with a mitochondrial basis has largely been developed by sequencing the mtDNA from persons showing signs of the disease.  But now there is the opportunity to try a different approach: one could discover what mitochondrial mutations exist and then consider what conditions they might cause.  The next few years can be expected to be very exciting.

 

 

Conclusion

 

In this article the medical implications of complete mitochondrial sequencing have been discussed;  and, the conclusion must be that for a fit person there is a most definite risk to submitting to complete mtDNA sequencing.  This risk is not high, with there being, maybe, a 5-10% chance of potentially harmful mutations being discovered.  But mitochondrial diseases are important and may well be much more common than hitherto realized, especially in association with non-specific illnesses.

 

 

Electronic-Database Information

 

Mitomap            http://www.mitomap.org

The main database for mitochondrial diseases and mutations

 

mtDB                 http://www.genpat.uu.se/mtDB/

A searchable database of published mtDNA sequences

 

Pubmed             http://www.ncbi.nlm.nih.gov/entrez/Query.fcgi?=db=pubmed

An important database for medical and scientific papers

 

NCBI                 http://www.ncbi.nlm.nih.gov/entrez/Viewer.fcgi?db=nucleotide

The main source for published mtDNA sequences

 

Mamit-trna         http://mamit-trna.u-strasbg.fr/index.html

                        A database of mammalian tRNAs

 

Software

 

MitoAnalyzer.         National Institute of Standards and Technology, Gaithersburg, MD, USA.

http://www.cstl.nist.gov/biotech/strbase/mitoanalyzer.html, 2000.

A useful program that shows the effect of mtDNA mutations

 

References

 

Achilli A, Rengo C, Magri C, Battaglia V, Olivieri A, Scozzari R, Cruciani F, Zeviani M, Briem E, Carelli V, Moral P, Dugoujon JM, Roostalu U, Loogväli EL, Kivisild T, Bandelt HJ, Richards M, Villems R, Santachiara-Benerecetti AS, Semino O, Torroni A (2004)  The molecular dissection of mtDNA haplogroup H confirms that the Franco-Cantabrian glacial refuge was a major source for the European gene pool.  Am J Hum Genet, 75:910-918.

 

Anderson S, Bankier AT, Barrell BG, de Bruijn MH, Coulson AR, Drouin J, Eperon IC, Nierlich DP, Roe BA, Sanger F, Schreier PH, Smith AJ, Staden R, Young IG (1981)  Sequence and organization of the human mitochondrial genome. Nature, 290:457-65.

 

Coble MD, Just RS, O'Callaghan JE, Letmanyi IH, Peterson CT, Irwin JA, Parsons T (2004)  Single nucleotide polymorphisms over the entire mtDNA genome that increase the power of forensic testing in Caucasians. Int J Legal Med, 118:137-146.

 

Janke A., Xu X., Arnason U (1997)  The complete mitochondrial genome of the wallaroo (Macropus robustus) and the phylogenetic relationship among Monotremata, Marsupialia and Eutheria.  Proc Natl Acad Sci USA, 94:1276-1281.

 

Kivisild T, Shen P, Wall D, Do B, Sung R, Davis K, Passarino G, Underhill PA, Scharfe C, Torroni A, Scozzari R, Modiano D, Coppa A, de Knijff P, Feldman M, Cavalli-Sforza LL, Oefner PJ (2005) The role of selection in the evolution of human mitochondrial genomes. Genetics, in press.

 

Macaulay V, Hill C, Achilli A, Rengo C, Clarke D, Meehan W, Blackburn J, Semino O, Scozzari R, Cruciani F, Taha A, Shaari NK, Raja JM, Ismail P, Zainuddin Z, Goodwin W, Bulbeck D, Bandelt HJ, Oppenheimer S, Torroni A, Richards M (2005)  Single, rapid coastal settlement of Asia revealed by analysis of complete mitochondrial genomes. Science, 308:1034-1036.

 

Sykes B. (2001) The Seven Daughters of Eve. Bantam Press.

 

Taylor RW, Turnbull DM (2005)  Mitochondrial DNA mutations in human disease.  Nat Rev Genet, 6:389-402.

 

 

Selected Further Reading

 

Of particular interest:

 

Levinger L, Morl M, Florentz C (2004)  Mitochondrial tRNA 3' end metabolism and human disease.  Nucleic Acids Res, 32:5430-41.

 

Other papers of interest:

 

Gattermann N, Wulfert M, Junge B, Germing U, Haas R, Hofhaus G (2005)  Ineffective hematopoiesis linked with a mitochondrial tRNA mutation (G3242A) in a patient with myelodysplastic syndrome Blood, 103:1499-1502.

 

Hao R, Zhao MW, Hao ZX, Yao YN, Wang ED (2005)  A T-stem slip in human mitochondrial tRNALeu(CUN) governs its charging capacity. Nucleic Acids Res, 33:3606-3613.

 

Jacobs HT, Hutchin TP, Kappi T, Gillies G, Minkkinen K, Walker J, Thompson K, Rovio AT, Carella M, Melchionda S, Zelante L, Gasparini P, Pyykko I, Shah ZH, Zeviani M, Mueller RF (2005)  Mitochondrial DNA mutations in patients with postlingual, nonsyndromic hearing impairment. Eur J Hum Genet, 13:26-33.

 

Kelley SO, Steinberg SV, Schimmel P (2000)  Functional defects of pathogenic human mitochondrial tRNAs related to structural fragility. Nat Struct Biol, 7:862-865.

 

Kelley SO, Steinberg SV, Schimmel P (2001)  Fragile T-stem in disease-associated human mitochondrial tRNA sensitizes structure to local and distant mutations. J Biol Chem 276:10607-10611.

 

Levinger L, Oestreich I, Florentz C, Morl M (2004)  A pathogenesis-associated mutation in human mitochondrial tRNALeu(UUR) leads to reduced 3'-end processing and CCA addition.  J Mol Biol, 337:535-544.

 

Moilanen JS, Finnilä S, Majamaa K (2003) Lineage-specific selection in human mtDNA: lack of polymorphisms in a segment of MTND5 gene in haplogroup J Mol Biol Evol, 20:2132-2142.

 

Ruppert V, Nolte D, Aschenbrenner T, Pankuweit S, Funck R, Maisch B (2004)  Novel point mutations in the mitochondrial DNA detected in patients with dilated cardiomyopathy by screening the whole mitochondrial genome.  Biochem Biophys Res Commun, 318:535-543.

 

Shin WS, Tanaka M, Suzuki J, Hemmi C, Toyo-oka T (2000)  A novel homoplasmic mutation in mtDNA with a single evolutionary origin as a risk factor for cardiomyopathy. Am J Hum Genet, 67:1617-1620.

 

Wilson FH, Hariri A, Farhi A, Zhao H, Petersen KF, Toka HR, Nelson-Williams C, Raja KM, Kashgarian M, Shulman GI, Scheinman SJ, Lifton RP (2004)  A cluster of metabolic defects caused by mutation in a mitochondrial tRNA.  Science. 306:1190-1194.