EXPERIMENTAL AND THEORETICAL STUDIES OF THE MITOCHONDRIAL
REPLISOME
By
Gregory Alan Farnum

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Biochemistry and Molecular Biology - Master of Science
2013

ABSTRACT
EXPERIMENTAL AND THEORETICAL STUDIES OF THE MITOCHONDRIAL
REPLISOME
By
Gregory Alan Farnum
Despite the ubiquitous trend of increasing complexity in eukaryotes, the vital process of
mtDNA replication has retained a simplistic, three component replisome consisting of the mtDNA
Pol γ, mtSSB and the mtDNA replicative helicase. This minimal mitochondrial replisome
resembles the bacteriophage T7 replisome and may even be simpler due to the apparent absence
of a primase function in the N-terminal domain of the mtDNA replicative helicase. Interestingly,
insects have retained the four cysteines of motif I which have been shown to bind zinc in T7, so
an N-terminal construct (NTD) of Drosophila melanogaster mtDNA replicative helicase was
designed containing the conserved residues. Purified fractions of the NTD construct contain an
iron sulfur cluster, as determined by UV-Visible spectroscopy, as well as iron and sulfide
determination. This work describes for the first time an iron sulfur cluster site in dm-mtDNA
replicative helicase and evidence of weak DNA binding by the NTD using a FRET assay.
Mutations in Pol γ represent a major cause of human mitochondrial diseases, especially those
affecting the nervous system in adults and in children. More than 160 POLG1 disease mutations
have been identified, which are nearly uniformly distributed along the length of the POLG1
sequence. Comprehensive literature analysis of the biochemical properties and disease
characteristics of Pol γ in light of the crystal structure have revealed genotype-phenotype
correlations that support the clustering of mutations into five functional modules in the catalytic
core of Pol γ. Our results suggest that cluster prediction can be used to evaluate both the likely
biochemical defects and the relative pathogenicity of new POLG variants.

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................................... vi
LIST OF FIGURES ...................................................................................................................... vii
KEY TO SYMBOLS AND ABBREVIATIONS ........................................................................ xiv
CHAPTER 1: The N-terminal primase-like domain of the Drosophila replicative mitochondrial
DNA helicase binds an iron sulfur cluster. ..................................................................................... 1
 INTRODUCTION .................................................................................................................... 2
 RESULTS ................................................................................................................................. 5
 Design of N-terminal truncation variants guided by T7 gp4 homology ....................... 5
 Purification of the NTD Dm helicase construct ............................................................ 6
 Dark brown fractions of NTD tend to aggregate without sufficient glycerol content and
thiol reductant. .............................................................................................................. 6
 Observation-NTD forms SDS and DTT resistant dimers and higher multimers in SDSPAGE gels. .................................................................................................................... 7
 The ZBD construct co-purifies with a similar color that is significantly lighter than the
NTD fractions. .............................................................................................................. 8
 Partial digestion patterns suggest the NTD has a modular architecture consisting of a
ZBD and RPD, similar to T7 gp4. ................................................................................ 9
 Electronic absorption spectra of purified Dm helicase fractions .................................. 9
o Optical features of as isolated NTD helicase samples and comparison to ZBD
spectra ............................................................................................................. 12
o Spectral features under reducing conditions ................................................... 12
o Spectral features of in vitro reconstituted samples ......................................... 14
o Stability of iron-sulfur cluster in the presence of hydrogen peroxide ............ 15
o Comparison to literature examples of iron sulfur cluster proteins .................. 16
 Chemical analysis of colored fractions by Graphite Furnace Atomic Absorption
Spectroscopy ............................................................................................................... 17
o Iron content ..................................................................................................... 18
o Zinc content .................................................................................................... 18
 Sulfide determination by methylene blue colorimetric method .................................. 19
 Electron paramagnetic resonance studies ................................................................... 19
o Instrumental interference obscured high field signals .................................... 21
o Low field signals may indicate cluster degradation upon reduction ............... 21
 Forster resonance energy transfer (FRET) experiments ............................................. 23
 Proposal and design for mutagenesis of the NTD construct ....................................... 26
o Experimental approach and possible outcomes .............................................. 29
 DISCUSSION ......................................................................................................................... 30
o Conclusions ..................................................................................................... 32

iii

 METHODS ............................................................................................................................. 33
 Bacterial cell growth and protein overexpression ....................................................... 33
 Cell lysis and preparation of soluble fraction (FrI) ..................................................... 33
 Ni-NTA metal affinity chromatography (FrII) ........................................................... 33
 Glycerol gradient sedimentation (FrIII) ...................................................................... 34
 Graphite furnace atomic absorption analysis of iron content of NTD (or ZBD) purified









fractions....................................................................................................................... 34
Quenching of fluorescence of fluorescein-labeled ssDNA (ssDNA-fl) by NTD via
FRET ........................................................................................................................... 35
o Calculate dissociation constant Kd for fl-dna binding by NTD ..................... 36
Methylene Blue Photometric Analysis of Sulfide Content in Protein Samples.......... 36
Inhibition of FRET quenching by salt ........................................................................ 37
Competitive FRET binding assay for polynucleic substrates ..................................... 37
o Calculate the inhibition coefficient Ki for each substrate .............................. 38
EPR Spectroscopy....................................................................................................... 39
UV-Visible Spectroscopy ........................................................................................... 39
Methods for mutagenesis ............................................................................................ 39
o Growth and preparation of Ca2+ competent cells ........................................... 39

Transformation of Ca2+ competent cells ........................................................ 40
Preparation of electrocompetent bacteria (XL-1 Blue)................................... 40
Mini-preparation of plasmid DNA: pET-28a- Dm mt helicase N24-A333 from
BL21(DE3) cells ............................................................................................. 41
o Transformation of pET-28a-dm mt helicase N24-A333 plasmid into XL-1 Blue
cells via electroporation .................................................................................. 41
o Mini-preparation of plasmid DNA: pET-28a- Dm mt helicase N24-A333 from
XL-1 Blue cells ............................................................................................... 42
o Quick change mutagenesis of dm mtDNA helicase N24-A333...................... 42
 APPENDIX ............................................................................................................................. 45
 BIBLIOGRAPHY ................................................................................................................... 51
o
o
o

CHAPTER 2: Clustering of Alpers disease mutations and catalytic defects in biochemical
variants reveal new features of molecular mechanism of the human mitochondrial replicase Pol
γ. .................................................................................................................................................... 55
 ABSTRACT............................................................................................................................ 56
 INTRODUCTION .................................................................................................................. 57
 Approach to comparative structural analysis .............................................................. 59
 Clustering of Alpers disease mutations within the catalytic core of Pol γ.................. 60
 RESULTS ............................................................................................................................... 64
 ALPERS CLUSTER 1: Residues affecting 5′-3′ DNA polymerase activity .............. 64
o Catalytic residues ............................................................................................ 64
o Architectural residues ..................................................................................... 67
o Residues conferring a high affinity of the pol site for the primer–template
junction. ......................................................................................................... 68
 CLUSTER 2: Recessive mutations affecting the upstream DNA-binding channel ... 72

iv



CLUSTER 3: Mutations associated with a novel, Pol γ-specific functional module,
conferring partitioning of DNA substrate between the pol and exo active sites ........ 74
 CLUSTER 4: Mutations affecting Pol γA interactions with the distal Pol γB upon DNA
binding by Pol γ holoenzyme ...................................................................................... 79
 CLUSTER 5: Mutations affecting a region of the IP subdomain that is likely involved
in replisome contacts................................................................................................... 82
 PROSPECTS........................................................................................................................... 85
 Toward a diagnostic tool to assess new human mitochondrial disease mutations ..... 85
 Predicting the consequences of compound heterozygosity via Alpers cluster analysis
..................................................................................................................................... 86
 APPENDIX ............................................................................................................................. 91
 BIBLIOGRAPHY ................................................................................................................... 96
CHAPTER 3: Mapping 136 Pathogenic Mutations into Functional Modules in Human DNA
Polymerase γ Establishes Predictive Genotype-phenotype Correlations for the Complete
Spectrum of POLG Syndromes. ................................................................................................. 100
 ABSTRACT.......................................................................................................................... 101
 INTRODUCTION ................................................................................................................ 102
 METHODS ........................................................................................................................... 106
 Computational analysis ............................................................................................. 106
 Statistical analysis ..................................................................................................... 106
 RESULTS ............................................................................................................................. 107
 Cluster assignment of all reported POLG disease mutations.................................... 107
 Cluster assignment can be used to predict the pathogenicity of a novel mutation. .. 116
 Cluster combinations of compound heterozygotes correlate with age of onset of POLG
syndrome. .................................................................................................................. 119
 Critical functions: clusters 1, 3, and 4....................................................................... 121
 Moderately severe dysfunction: clusters 2 and 5 ...................................................... 123
 Severe combination: cluster 1 + 2 ............................................................................. 124
 Severe combination: cluster 1+5 ............................................................................... 125
 Other combinations ................................................................................................... 125
 Expected incidence of specific symptoms caused by POLG syndrome can be predicted
by the age of onset. ................................................................................................... 126
 DISCUSSION ....................................................................................................................... 130
 APPENDIX ........................................................................................................................... 134
 BIBLIOGRAPHY ................................................................................................................. 146

v

LIST OF TABLES

Table 1. A comparison of UV-visible electronic absorption spectra of 2Fe-2S and 4Fe-4S
clusters from published studies. The UV-visible spectra collected for the NTD construct (on
page 29) show a peak at 280 nm corresponding to aromatic amino acids in the protein, a shoulder
at about 333 nm, and a peak at about 425 nm.
17
Table 2. Calculated ratios of Fe: protein for the helicase and cyt c samples from graphite
furnace atomic absorption data. Sample concentration was determined by the absorbance at
280 nm for purified helicase variants and by Bradford for the cyt c sample. All samples were
measured in triplicate. Error bars were calculated by propagated error analysis and all
measurements, dilutions, etc were assigned an uncertainty that was appropriate for the
measurement, instrument, pipette, etc. (ex: my P20 pipette was calibrated and found to have a
standard deviation of 0.6 µL, so any measurement or transfer with this pipette, was given an
uncertainty equivalent to the standard deviation (ex: 9.0 ± 0.6 µL).
19
Table 3. Physical properties based on primary sequence of NTD helicase. Theoretical
physical properties as determined from amino acid sequence of Dm mtDNA helicase N24A333.

45

Table 4. Protein content and purity of purification steps. Total protein concentration was
determined by A280 analysis and by the colorimetric Bradford method throughout purification.
Purity was estimated by SDS-PAGE analysis of target band versus contaminant bands.
46
Table 5. Structural and functional features of the five proposed Alpers Clusters.
Table 6. Grouping of POLG syndrome symptoms

vi

84
129

LIST OF FIGURES

Figure 1. Human minimal mitochondrial replisome. Orange arrow with dashed lines
illustrates leading strand synthesis by human DNA polymerase γ. Three different colors
represent domains of the mtDNA helicase that are discussed throughout the thesis: Purple, Cterminal helicase domain; Blue, RNA polymerase domain (RPD); Brown, Zinc binding domain
(ZBD); Blue + Brown, equivalent to the N-terminal domain (NTD) construct. POLGA, human
DNA polymerase γ catalytic subunit; POLGB, human DNA polymerase γ accessory subunit;
mtSSB, human mitochondrial single-stranded DNA-binding protein; Twinkle, human
mitochondrial replicative DNA helicase. “For interpretation of the references to color in this and
all other figures, the reader is referred to the electronic version of this thesis.”
4
Figure 2. Schematic representation of Dm helicase domain motifs NTD (36 kDa) and ZBD
(12 kDa) constructs, and multiple sequence alignment of the zinc binding Motif I. Arrows
indicate positions of four conserved cysteine residues that ligate zinc in T7 primase.
5
Figure 3. The NTD has a modular architecture consisting of a ZBD and RPD, similar to T7
gp4. SDS-PAGE (right) and anti-his immunoblot (left) analysis of 36 kDa N24-A333 construct
after limited trypsin digestion.
9
Figure 4. UV-visible electronic absorption spectrum of NTD. Peaks at 422 nm, 464 nm, and
590 nm suggest the presence of an iron-sulfur cluster by comparison with published examples of
UV-visible electronic absorption spectra of protein bound 2Fe-2S and 4Fe-4S clusters.
11
Figure 5 Comparison of UV-visible spectrum of NTD vs ZBD purified fractions. Pink trace
represents NTD spectrum, both red and green are ZBD fractions after Ni-NTA fractionation. 11
Figure 6. UV-Visible spectrum of the NTD construct after treatment with a strong
reducing agent. An immediate bleaching effect of dithionite on as isolated NTD Dm helicase
sample (blue) after addition of 8 fold excess dithionite (red). Change in volume was negligible
(2 µL in 500 µL) compared to the observed change in absorption at 420 nm.
13
Figure 7. Effect of DTT on the NTD Dm helicase sample after 36 hours on ice (red). As
isolated NTD (blue) is less intense in 300-325nm range. Concentration of as isolated NTD Dm
helicase is 28.5 μM. The oxidation of DTT to the cyclic configuration is accompanied by
increased absorption below 270-340 nm.
14
Figure 8. UV-Visible analysis of in vitro reconstitution of iron sulfur clusters. The bottom
most (pink) trace represents 1:1 ferrous iron and sulfide: helicase immediately after addition of
NTD helicase (28.5 μM). The second trace (green) is after 1 hour, the third trace (red) is after
addition of 2:1 ferrous iron and sulfide: helicase, and the top trace (blue) is after 12 hours.
15
Figure 9. The Fe-S cluster is resistant to oxidizing conditions. Normalized absorbance plot
for hydrogen peroxide titration of NTD Dm helicase. Blue: 7 μM as isolated NTD Dm helicase,
Red: 10:1 H2O2: helicase, Green: 100:1 H2O2: helicase.
16

vii

Figure 10. Low field EPR spectrum of as isolated NTD Dm helicase. Performed by a student
prior to my time in the lab, sample contains EDTA. High field signals near g = 2 are obscured
by a known copper contaminant in the instrument (see appendix on page 63).
22
Figure 11. Low field EPR spectrum of NTD Dm helicase after the addition of dithionite.
Performed by a student prior to my time in the lab, sample contains EDTA. High field signals
near g = 2 are obscured (see appendix on page 63).
22
Figure 12. Low field EPR spectrum of NTD Dm helicase after the addition of ferricyanide.
Performed by a student prior to my time in the lab, sample contains EDTA. High field signals
near g = 2 are obscured (see appendix on page 63).
23
Figure 13. Quenching of fluorescence in the presence of NTD construct. Excitation of
Fluorescein (red)-labeled ssDNA (ssDNA-fl) in the absence of NTD exhibited normal
fluorescence and addition of purified NTD construct (blue) resulted in quenching of
fluorescence. Iron sulfur clusters (brown star) can act as acceptors in FRET and result in
quenching.

24

Figure 14. Increasing ionic strength inhibits the quenching observed by the FRET assay.
Salt ions can compete with the ionic interactions at the interface of DNA – protein complexes
and increase the rate of dissociation and inhibit re-binding. Addition of salt (NaCl and
MgCl2).to a solution containing 10 nM fl-ssDNA/ 5 uM NTD Dm helicase caused an immediate
increase in fluorescent intensity. By inhibiting the binding of fl-ssDNA to the NTD the
quenching of fluorescence by FRET is prevented because the Fe-S cluster is no longer in close
proximity to the fluorescein. Complete inhibition of FRET quenching is observed at 150 mM
NaCl and 70 mM MgCl2.
25
Figure 15. FRET assay shows that quenching can be inhibited competitively by unlabeled
ssDNA. Semi-log plots of the FRET competitive binding assay comparing the polydT substrate
with ssRNA substrate, polyrA. Concentrations of DNA and RNA substrates are by nucleotide
concentration because the length of individual molecules are not constant.
26
Figure 16. Alignment of fly, frog and human mtDNA helicases for mutagenesis design and
selection of conserved cysteine residues. Cysteines highlighted in green will be produced as
single alanine substitutions in the NTD construct and cysteines highlighted in cyan will be
produced as double alanine substitutions for each pair of CXXC motifs. Red blocks (α-helical)
and blue arrows (β-sheet) below the sequence illustrates the predicted secondary structure. 28
Figure 17. SDS-PAGE analysis of NTD construct throughout the purification. We suspect
an SDS-resistant disulfide-linked dimer is the band at 70 kDa.
46
Figure 18. Gel filtration analysis of NTD construct using a Superdex 75 column. The
elution of a single peak corresponding to a monomer as determined by the A280 measurements
shown as the blue trace.
47
Figure 19. Instrument interference at g = 2 prevented further characterization of Fe-S
characteristics by EPR. This EPR spectrum was taken with an empty sample tube to show that
the large signal at g = 2 is suspected to be a copper impurity in the instrument and not due to
sample contamination.
47

viii

Figure 20. Example of semi-log plot and fit used to calculate the IC50. (The x axis is
logarithmic).

48

Figure 21. Example UV-Visible spectrum for standards: The reaction is considered
successful if the ratio of absorbance between A670 and A750 is close to 2 as is observed in these
spectra. Blue- .5 nmol sulfide, Red- 1 nmol sulfide, Pink- 2 nmol sulfide, Dark green- 3 nmol
sulfide, Brown- 4 nmol sulfide.
49
Figure 22. Alpers disease mutations cluster within functional modules in the catalytic
subunit of Pol γ. Upper panel: schematic diagram of the POLG1 gene showing the distribution
of recessive Alpers disease mutations (Human DNA Polymerase gamma Database,
http://tools.niehs.nih.gov/ polg/). AID and IP in the spacer domain refer to the accessory
(subunit) interacting and intrinsic processivity subdomains, respectively, that are discussed in the
text; NTD refers to the N-terminal domain. Lower panel: tertiary structural representation of the
apoenzyme form of Pol γ [PDB code 3IKM, (7)] with modeled DNA, identifying the positions of
five functional modules (shown in mesh) that are defined by clusters of amino acid residues
(shown as spheres) affected by Alpers disease mutations as follows: Cluster 1, green; Cluster 2,
yellow; Cluster 3, red; Cluster 4, blue; Cluster 5, cyan. The domains of Pol γA are shown as
surface representations, and in part as secondary structural elements (SSEs) that are colored as
depicted in A. The proximal and distal accessory subunits are shown as surface representations
in light and dark gray, respectively. Primer–template DNA was docked as described in the text
and is displayed as orange ribbons.
62
Figure 23. Alpers Cluster 1 mutations affect the 5 DNA polymerase activity of Pol γ.
Amino acid residues affected by Alpers Cluster 1 mutations in POLG1 are shown as green
spheres. Other Pol γ residues that are discussed in the text are shown in brown, and T7 residues
are shown in red. Pol domain SSEs are shown in pink according to the schematic shown in
Figure 22 on page 77; ptDNA is indicated by orange (template) and brown (primer) strands. The
incoming dNTP is shown in blue. Mg2+ ions are shown as small gray spheres. A, upper panel,
overview of the positions of Alpers Cluster 1 mutations with dashed black lines indicating the
regions described in the text and depicted in B–D; A, lower panel, overview of the positions of
Alpers Cluster 1 mutations relative to ptDNA and incoming dNTP; B–D, positioning of Pol γ
residues in the apoenzyme form [PDB code 3IKM (7)] and T7 Pol residues in its closed ternary
complex [PDB code 1T8E (25)] relative to ptDNA and incoming dNTP, with the dashed arrow
showing the expected movements of the Pol γ residues upon formation of a closed complex: B,
Alpers mutations surrounding the Mg2+ -binding residues; C, Alpers mutations affecting O-helix
movement and dNTP binding, with the T7 Pol O-helix in its closed conformation superimposed
in red; D, Alpers mutations affecting the RR loop and the surrounding residues.
70
Figure 24. Alpers Cluster 2 mutations affect the upstream DNA-binding channel of Pol γ.
Amino acid residues affected by Alpers Cluster mutations in POLG1 are shown as yellow
spheres. Other Pol residues that are discussed in the text are shown in brown. Spacer domain
SSEs are shown in magenta and pol domain SSEs are in pink according to the schematic shown
in Figure 22 on page 77; ptDNA is indicated by orange (template) and brown (primer) strands.
The spacer domain is also shown as a transparent surface representation in pale gray and the exo
domain is shown in purple.
74

ix

Figure 25. Alpers Cluster 3 mutations are associated with a novel Pol γ-specific functional
module proposed to be involved in primer strand partitioning between the pol and exo
sites. Amino acid residues affected by Alpers Cluster 3 mutations in POLG1 are shown as red
spheres/ mesh adjacent to a novel alpha helix with an associated loop–hairpin (the partitioning
loop), also shown in red. The brown spheres and mesh represent the SYW (fly)/ SFW (human)
and surrounding residues, respectively, that are described in the text. The pol domain is shown as
a surface representation in pink and the exo domain is shown in purple, according to the
schematic shown in Figure 22 on page 77; ptDNA is indicated by orange (template) and brown
(primer) strands. (A) The predicted position of the partitioning loop relative to ptDNA in the pol
mode, and (B) represents the exo mode. To dock the frayed ptDNA in the exo active site, the exo
domain residues 324–518 of Klenow (PDB code 1KLN) were aligned with the exo domain
residues 170–440 of Pol γ (PDB code 3IKM) (see appendix B for Figure 31 on page 105).
78
Figure 26. Alpers Cluster 4 mutations affect Pol γA interactions with the distal Pol γB
upon DNA binding by Pol γ holoenzyme. Amino acid residues affected by Alpers Cluster 4
mutations in POLG1 are shown as blue spheres/ mesh and are located largely within the exo
domain (shown in purple). The helix shown in magenta is the AID (accessory interacting
domain) of the spacer region that interacts with the proximal accessory subunit. Other domains
are colored according to the schematic shown in Figure 22 on page 77; duplex DNA is indicated
by the orange strand. Orange spheres indicate the positions of the accessory subunit residues
described in the text. The proximal and distal accessory subunits are shown as surface
representations in light and dark gray, respectively.
81
Figure 27. Alpers Cluster 5 mutations are proposed to affect replisome interactions. Amino
acid residues affected by Alpers Cluster 5 mutations in POLG1 are shown as cyan spheres.
Spacer domain SSEs within the IP (intrinsic processivity subdomain) are shown in magenta;
ptDNA is indicated by orange (template) and brown (primer) strands. The spacer domain is also
shown as a transparent surface representation in pale gray and the exo domain is shown in
purple. Brown spheres indicate the positions of the IP residues described in the text.
83
Figure 28. Combinations of mutations found in Alpers patients. Individual Alpers mutations
are grouped by cluster as shown in Figure 22 on page 77, and black blocks represent Alpers
manifesting combinations. The inset in the lower right presents a simplified version of the table
by reducing the axes to the five clusters only, where gray blocks represent cluster combinations
that have not been found in Alpers patients. The tabulated data suggest that two mutations from
the same cluster do not typically manifest as early-onset Alpers disease. Furthermore, the data
indicate that Cluster 4 (blue) mutations manifest as Alpers disease only when in combination
with Cluster 2 (yellow) or Cluster 5 (cyan) mutations. These trends support the existence of
unique functional relationships in Pol γA that are inherent to each cluster.
88
Figure 29. Alignments used for docking DNA onto the Hs Pol γ holoenzyme structure
(3IKM). Panels display the palm subdomain of the T7 Pol (1T8E) closed ternary complex
(red) superimposed on the palm subdomain of Pol γ (pink). Top panel, alignment of the palm
of T7 Pol (residues 409-487, 611-704) to the palm of Pol γ (815-910, 1095-1239); middle panel,
alignment of T7 Pol residues 646-663 (green) to Pol γ residues 1127-1144; lower panel,
alignment of T7 Pol residues 606-635 and 422-431 (green) to Pol γ residues 1093-1122 (Q-helix)
and 846-855 (RR-loop). Alignments were performed in Pymol.
91

x

Figure 30. Comparative alignments of Hs Pol γ, T7 Pol and T7 RNA Pol illustrating the
structural variations in the region between the P-helix and the Q-helix in the pol domain.
Panels on the left display the complete pol domain in complex with DNA, and those at right
show a close up view of the of the region between the Q-helix and P-helix. Top panel, the Pol γ
(3IKM) pol domain is shown as pink cartoon and its partitioning loop in red, with its disordered
region indicated as a dashed red line. Transparent surfaces of the exo (purple) and pol (pink)
domains are also shown at left. DNA docked by alignment with T7 Pol (see top panel of Fig. 29
on page 103) is shown in orange; middle panel, the Pol domain of the T7 RNA Pol elongation
complex (1H38) is shown as light blue cartoon and its specificity loop in red; lower panel, the T7
Pol (1T8E) pol domain is shown as blue cartoon, and transparent surfaces of its exo (purple) and
pol (blue) domains are shown to highlight a similar architecture as in Pol γ. WXGG and WXAG
are the amino acid sequence motifs in Pol γ and T7 Pol, respectively, which are discussed in the
text.
92
Figure 31. Structural alignment of the exo domain of Klenow editing complex (PDB code
1KLN, residues 324-518, displayed as blue cartoon) with the exo domain of human Pol γ
(PDB code 3IKM, residues 170-440, displayed as purple cartoon). This alignment was used
to predict the editing complex of Pol γ by docking of the frayed primer template (primer strand
in chocolate, template strand in orange) onto the apoholoenzyme structure, which we display in
Figure 25B on page 91.
94
Figure 32. Clustering of 136 pathogenic mutations within five functional modules in the
catalytic subunit of human Pol γ. Upper panel, schematic diagram of the human POLG1 gene
illustrating the clustering of 136 pathogenic mutations into discrete blocks of amino acid
residues, which we term subclusters. Mutant alleles and subclusters are colored according to the
cluster to which they belong: cluster 1, green; cluster 2, yellow; cluster 3, red; cluster 4, blue;
cluster 5, cyan (see the text for details). The palm (residues 815-910 and 1096-1239), fingers
(residues 911-1048), and thumb (Th, residues 440-475 and 785-814) subdomains of the Pol
domain are colored pink, and the partitioning loop (PL, residues 1049-1095) is red. The
accessory (subunit) interacting domain (AID, residues 476-570) and the intrinsic processivity
(IP, residues 571-784) subdomains of the spacer domain are colored in magenta. The N-terminal
domain (NTD, residues 1-170) and the Exo domain (residues 171-439) are colored in purple.
Lower panel, structural model of the human Pol γ apo-holoenzyme (PDB code 3IKM) with
docked primer template DNA shown as orange ribbon (see Computational methods for details).
The catalytic subunit of Pol γ is shown as a cartoon representation of the secondary structural
elements (SSEs), with regions defined by clusters illustrated as space-filled modules, colored
according to the schematic. The proximal and distal accessory subunits are shown as surface
representations in light and dark gray, respectively.
110

xi

Figure 33. Architectural and functional subclusters of the Pol domain. Upper panel,
schematic diagram of the POLG1 gene as shown in Figure 1, with an additional upper section
indicating the location of the three Exo motifs, labeled as I, II, and III, and the six Pol motifs,
labeled as 1 (motif 1), 2 (motif 2), A (Pol A motif), B (Pol B motif), 6 (motif 6) and C (Pol C
motif), which are conserved throughout family A polymerases (see text). Motifs and subclusters
that are illustrated in the bottom panels are in shown bold. Bottom panels, SSEs and transparent
surface representation of the palm and fingers subdomains of Pol γ (PDB code 3IKM) are shown
(subdomains are colored in gray in the middle section of the schematic). Bottom-left panel, the
five motifs of the Pol domain, which are colored in pink in the upper section of the schematic,
are shown as pink SSEs and are labeled accordingly. Docked primer template DNA is shown as
orange ribbon, Mg2+ ions are shown as orange spheres, and the incoming dNTP is shown as
orange sticks (see Computational methods for details). Bottom-right panel, subclusters 1D, 1E
and 1F encompass one or more of the five conserved motifs of the Pol domain and are colored in
light green. Subclusters 1A, 1B, 1C and 1G, colored in dark green, are located further
structurally from the pol active site, and are considered to play an architectural role.
112
Figure 34. SNPs reported in the POLG1 gene rarely map to clusters. Of 87 reported SNPs in
the dbSNP database, 75 (86%, black boxes) map outside the defined pathogenic clusters (shown
as colored regions in the schematic). The remaining 12 SNPs (magenta boxes) that map within
the clusters can be considered as non-deleterious changes within a high-risk region of the POLG
tertiary structure. However, in our view, the possible pathogenicity of the SNPs reported in
cluster 4 (dark blue) warrants careful experimental reevaluation.
118
Figure 35. Analysis of mutation combinations as cluster combinations reveals predictive
genotype-phenotype correlations. The data used to compile the information in this figure was
derived from the literature listed under supplemental references in appendix C. The number of
mutation combinations manifesting POLG syndrome at each of the four age groups is shown in
each panel for specific cluster combinations. Age of onset trends can be used to predict the
severity of POLG syndrome for an individual with compound heterozygous mutations in
POLG1. Upper panel, POLG syndrome age of onset trends for mutation combinations of two
mutations from different clusters versus the same cluster. Middle panel, patient data show trends
in which severe cluster combinations have an earlier age of onset, whereas less severe or
uncommon cluster combinations have a later age of onset. Lower panel, earlier age of onset
trends for more severe cluster/ subcluster combinations.
120
Figure 36. Analysis of all compound heterozygous mutation combinations as cluster
combinations. The data used to compile the information in this figure was derived from the
literature listed under supplemental references in appendix C. The number of mutation
combinations manifesting POLG syndrome at each of the four age groups is shown for all cluster
combinations for which patient data have been reported; no patient data have been reported for
the three cluster combinations 3+4, 4+4 and 4+5. Age of onset trends can be used to predict the
severity of POLG syndrome for an individual with compound heterozygous mutations in
POLG1. Patient data show trends in which severe cluster combinations have an earlier age of
onset, whereas less severe or uncommon cluster combinations have a later age of onset.
122

xii

Figure 37. Age of onset correlation for reported symptoms. Symptoms associated with
POLG syndromes manifest typically at different ages. The symptoms reported in the patients
have been grouped together based on affected tissue types, and ordered according to their ages of
onset. The more severe symptoms and forms of POLG syndromes are at the left end of the
figure, while the less severe symptoms are at the right end. Detailed information of the symptom
grouping is shown in Table 6.
127
Figure 38. Symptoms associated with different POLG syndromes. Depending on the specific
mutations and cluster combinations of the POLG1 gene mutations in individual patients, POLG
syndromes can become symptomatic at different ages and manifest as pathogenic conditions in
different tissue types. The figure shows a continuing spectrum of decreasing symptom severity
from top to bottom, as well as a delayed age of onset from left to right. MCHS, Childhood
myocerebrohepatopathy spectrum; MEMSA, Myoclonic epilepsy myopathy sensory ataxia;
ANS, Autonomic Nervous System Dysfunction; arPEO/ adPEO, Autosomal Recessive/
Dominant Progressive External Ophthalmoplegia.
128

xiii

KEY TO SYMBOLS AND ABBREVIATIONS

2Fe-2S

2 iron 2 sulfur cluster

3Fe-4S

3 iron 4 sulfur cluster

4Fe-4S

4 iron 4 sulfur cluster

6x His

6x histidine tag

Å

angstrom

aa

amino acid

Ab

antibody

adPEO

autosomal dominant progressive external ophthalmoplegia

anti-his

anti 6x histidine tag antibody

ATP

adenosine triphosphate

cm-1 M-1

extinction coefficient

cyt c

horse heart cytochrome c

Dm

Drosophila melanogaster

Dm helicase

Drosophila melanogaster mitochondrial replicative DNA helicase

DMPD

N, N-dimethyl-p-phenylenediamine

DNA

deoxyribonucleic acid

dsDNA

double-stranded DNA

DTT

dithiothreitol

E. coli

Escherichia coli

EDTA

ethylene diamine tetra acetic acid

xiv

EPR

electron paramagnetic resonance

Exo

3’→5’ exonuclease

FA

fluorescence anisotropy

Fe-S

iron sulfur cluster

fl-ssDNA

fluorescein-labeled ssDNA

FRET

Forster resonance energy transfer

GFAA

Graphite furnace atomic absorption

GHz

gigahertz

H2O2

hydrogen peroxide

HOMO

highest occupied molecular orbital

HsmtDNA helicase

Human mtDNA helicase

Hspol γ

Human DNA polymerase γ

kDa

kilodalton

ki

coefficient of inhibition

Klenow

bacterial DNA polymerase I large fragment

Klentaq

Thermus aquaticus Klenow

LUMO

lowest unoccupied molecular orbital

mg/ ml

milligram per milliliter

ms

angular spin momentum

mtDNA

mitochondrial DNA

mtDNA helicase

mitochondrial replicative DNA helicase

mtSSB

mitochondrial single-stranded DNA-binding protein

xv

Ni-NTA

nickel nitriloacetic acid

nm

nanometers

NTP

nucleoside triphosphate

NTPase

nucleoside triphosphatase

OXPHOS

oxidative phosphorylation system

Pol I

bacterial DNA polymerase I

Pol γ

DNA polymerase gamma

Pol γA

DNA polymerase gamma catalytic subunit

Pol γB

DNA polymerase gamma accessory subunit

Pol

5’→3’ DNA polymerase

POLG1

Pol γA gene

ppb

parts per billion

ptDNA

primer-template DNA

RNA

ribonucleic acid

RNAP

RNA polymerase basic domain

ROS

reactive oxygen species

RPD

RNA polymerase domain

SDS

sodium dodecyl sulfate

SDS-PAGE

sodium dodecyl sulfate polyacrylamide gel electrophoresis

SSB

single-stranded DNA-binding protein

ssDNA

single-stranded DNA

ssDNA-fl

fluorescein labeled ssDNA oligo used in FRET assays

SSE

secondary structural element

xvi

T7 gp4

bacteriophage T7 gene 4 protein

T7 gp5

bacteriophage T7 DNA polymerase

T7 Pol

bacteriophage T7 DNA polymerase

TOPRIM

topoisomerase-primase

UV-visible

Ultraviolet-visible (electronic absorption)

ZBD

zinc binding domain

μM

micromolar

xvii

CHAPTER 1: The N-terminal primase-like domain of the Drosophila replicative
mitochondrial DNA helicase binds an iron sulfur cluster.

1

INTRODUCTION
The mitochondrial replicative DNA helicase (mtDNA helicase), also known as Twinkle in
the human system, is an essential component of the mitochondrial replisome (1). The mtDNA
helicase catalyzes the NTP-dependent unwinding of DNA at the mitochondrial replication fork in
the 5'-3' direction. In vitro studies have established a minimal mitochondrial replisome consisting
of the mtDNA helicase and three additional purified protein components (Figure 1 on page 4): the
DNA polymerase gamma catalytic subunit (Pol γA), the DNA polymerase gamma accessory
subunit (Pol γB) and the mitochondrial single-stranded DNA-binding protein (mtSSB). Together,
these four components can generate DNA products greater than 16 kb in length, which is similar
in size to the human mitochondrial genome (1). Mutations affecting the PEO1 gene that encodes
the human mtDNA helicase have been reported in patients with adult-onset autosomal dominant
progressive external ophthalmoplegia (adPEO), as well as compound heterozygous patients with
more severe mitochondriopathies such as infantile-onset spinocerebellar ataxia and hepatocerebral
mtDNA depletion syndromes (1, 2).
The mtDNA helicase was identified originally for its sequence similarity to bacteriophage
T7 gp4 primase-helicase (T7 gp4). Biochemical investigation of human the mtDNA helicase
revealed a modular architecture that is similar to the T7 gp4 primase-helicase, as they both contain
distinct N and C-terminal domains separated by a flexible linker domain (1). In both proteins the
C-terminal domain harbors the active site responsible for NTPase activity, and can unwind duplex
DNA substrates independently. The C-terminal domain is considered to be in the DnaB-like
family of helicases, which form ring shaped hexamers that encircle the DNA substrate (3). Two
motifs that are well known are the Walker A and Walker B motifs that are involved in nucleotide
binding, and in all hexameric helicases these motifs are located at the subunit interface (3).
Binding of ATP allows interaction with an arginine finger residue of an adjacent monomer and
2

causes conformational shifts in four of six subunits. This asymmetric movement provides the 5'3' directional translocation along the ssDNA due to loops that extend into the center of the ring
from each monomer. ATP hydrolysis allows propagation of these movements for continuous
motion. Conserved motifs in the DnaB-like family are present in all known mtDNA helicase
amino acid sequences and have been reviewed previously (4). Overall, the C-terminal domain has
been well characterized and the focus of this thesis chapter is on the function of the N-terminal
domain.
The N-terminal domain of T7 primase-helicase consists of two functional domains; a RNA
polymerase domain (RPD) that synthesizes RNA primers to be extended by a DNA polymerase,
and a zinc binding domain (ZBD) that is involved in priming site recognition and stabilization of
the nascent primer-template complex to ensure efficient delivery to the replicative T7 DNA
polymerase (2). Whereas the N-terminal domain of Hs mtDNA helicase shows significant
sequence similarity to the N-terminal domain of T7 primase-helicase, no primase activity has been
reported in vitro. Therefore, the role of the N-terminal domain remains unclear, but is of interest
due to the evolutionary conservation of N-terminal sequence motifs in a wide range of eukaryotic
organisms from algae to fly to humans (4). Most studies on the mtDNA replicative helicase have
focused on the human enzyme, and biochemical studies have been limited due to solubility issues
during purification (5). Our lab has faced the same solubility issues with human mtDNA helicase
purifications and we have attempted to characterize the Drosophila melanogaster mtDNA helicase
(Dm Helicase) to possibly avoid similar solubility issues. This first thesis chapter is a report of
my initial attempts to study the mtDNA helicase from the Drosophila melanogaster system.

3

Figure 1. Human minimal mitochondrial replisome. Orange arrow with dashed lines
illustrates leading strand synthesis by human DNA polymerase γ. Three different colors
represent domains of the mtDNA helicase that are discussed throughout the thesis: Purple, Cterminal helicase domain; Blue, RNA polymerase domain (RPD); Brown, Zinc binding domain
(ZBD); Blue + Brown, equivalent to the N-terminal domain (NTD) construct. POLGA, human
DNA polymerase γ catalytic subunit; POLGB, human DNA polymerase γ accessory subunit;
mtSSB, human mitochondrial single-stranded DNA-binding protein; Twinkle, human
mitochondrial replicative DNA helicase. “For interpretation of the references to color in this and
all other figures, the reader is referred to the electronic version of this thesis.”

4

RESULTS
Design of N-terminal truncation variants guided by T7 gp4 homology
The N-terminal region of metazoan mtDNA helicase shares extensive sequence
conservation with the primase domain of bacteriophage T7 primase-helicase (4). Although several
Mg2+ binding residues of the primase active site in T7 primase are not conserved in metazoans,
the conserved primase sequence motifs are well conserved (6). Interestingly, insects have retained
the four cysteines of motif I that have been shown to bind zinc in T7 gp4 (Figure 2 below).
Therefore, to study the function of the N-terminal domain of metazoan mtDNA helicase, an Nterminal construct of Drosophila melanogaster mtDNA helicase (Dm helicase) was designed to
comprise residues N24-A333 NTD.

Based on T7 gp4 homology, this region contains the

conserved zinc binding domain (ZBD) and RNA-polymerase domain (RPD). In addition to the
NTD construct, a ZBD construct of Dm helicase that comprises residues N24-P123 was also
produced.

An

N-terminal

6x

histidine

tag

was

added

to

both

constructs.

Figure 2. Schematic representation of Dm helicase domain motifs NTD (36 kDa) and ZBD
(12 kDa) constructs, and multiple sequence alignment of the zinc binding Motif I. Arrows
indicate positions of four conserved cysteine residues that ligate zinc in T7 primase.
5

Purification of the NTD Dm helicase construct
E. coli cells containing IPTG-inducible plasmid with cDNA encoding Dm helicase ZBD
and NTD constructs were grown, induced, and harvested via centrifugation to produce pellets that
were frozen in liquid nitrogen and stored at -80º C until purification. Previous lab members had
completed several purifications of the NTD mutant derived from bacterial overexpression, and
they were able to obtain high yields only when the freeze-thaw lysis was performed in the presence
of DoDM instead of sodium cholate. Despite their initial success, they reported several failed
purifications due to spontaneous aggregation in performing chromatography steps at 4º C in the
presence of protease inhibitors, just as was found for the full-length human mtDNA helicase does
under similar conditions. Furthermore, the chromatographic columns that were used after the NiNTA fractionation step were unable to remove the apparent contaminants observed as larger
molecular weight bands on SDS-PAGE gels. Additionally, the protein eluted as a yellow solution
and would become progressively darker upon concentration. So the study of the NTD construct
was set aside and became my project about a year later. I set out first to document the origin of
the brown color.
Dark brown fractions of NTD tend to aggregate without sufficient glycerol content and
thiol reductant.
In all subsequent purifications we avoided the use of EDTA and other chelating agents in
purification buffers to avoid disrupting the native iron-sulfur cluster. Fresh β-mercaptoethanol
was replenished each day in all NTD-containing solutions or the NTD was observed to aggregate
irreversibly. These two small changes to the purification procedure improved noticeably the
stability of the NTD throughout purification, preventing aggregation of samples stored on ice
overnight. The prevention of aggregation showed the most severe dependence on a high glycerol

6

content of 10-15% in all solutions; we observed on several occasions the signs of aggregation after
only 15 min upon dilution.
Observation-NTD forms SDS and DTT resistant dimers and higher multimers in SDSPAGE gels.
As a result of the stabilization of NTD samples, purification of the NTD was achieved in
only two chromatographic steps (See appendix A on page 46). SDS-PAGE gels showed a nearhomogenous NTD species, but a 70 kDa species that had inconsistent intensity from gel to gel was
still present and needed to be explained (see appendix A on page 46). The last step of the
purification is a velocity sedimentation in a glycerol gradient, which can be used to estimate the
apparent molecular weight in addition to separating protein contaminants. The NTD construct
sediments at a position consistent with a monomer of 35 kDa. Consistent with the sedimentation
results, the NTD eluted as a sharp peak at a position corresponding to 35 kDa on a Superdex 75
gel filtration column (See appendix on page 47). Samples were golden-brown and became very
dark when the protein was concentrated. Analysis of the Ni-NTA fractions by SDS-PAGE
revealed the presence of SDS-resistant multimers of the target protein in the early eluting fractions
of the protein peak. To confirm that these bands were indeed multimers of the target protein and
not bacterial protein contaminants, an immunoblot of the fractions was performed using a primary
antibody that recognizes the 6x histidine tag. As suspected, the multimer bands were recognized
and stained by the anti-his antibody. We speculate that the high cysteine content of the construct
facilitates non-native, intermolecular disulfide bond formation during the SDS-denaturation
process. Analysis of the velocity sedimentation fractions by SDS-PAGE showed that fractions
with a greater rate of sedimentation exhibited SDS-resistant multimers. In addition, we observed
that the fractions that produced SDS-resistant multimers had a lesser intensity of brown coloration
as compared to fractions that did not produce multimers. We found subsequently that the ratio of
7

iron and sulfide to protein was significantly lower in the fractions that produced multimers, and
we speculate that the apo-protein is more likely to produce SDS-resistant disulfide-linked
multimers. Iron-sulfur clusters are ligated by four cysteine residues and in the apo form, these
cysteine residues would be free to form disulfide bonds adventitiously.
The ZBD construct co-purifies with a similar color that is significantly lighter than the
NTD fractions.
Purification of the ZBD construct was performed under the same conditions as the NTD
construct using a Ni-NTA agarose column. Similar to the NTD construct, a brown band was
observed on the Ni-NTA resin when the protein was loaded onto the column and eluted as goldenbrown fractions. A similar pattern of SDS-resistant multimers was observed for the early eluting
fractions, and they were determined to have a lower ratio of iron and sulfide to protein than
fractions which did not produce SDS-resistant multimers. The peak ZBD fractions exhibited
similar UV-visible spectra to the purified NTD. However, the highest ratio of iron and sulfide
atoms to protein observed was 0.9. In addition, the purified ZBD fractions were unstable at 4º C,
as evidenced by the appearance of significant aggregation after only 4 hours. Many fractions of
near-homogenous ZBD were observed to yield a ladder of bands as for the NTD, and were
confirmed to contain a 6x His tag by immunoblot. We conclude that the ZBD is able to bind the
iron sulfur cluster, but the binding is less stable than in the NTD construct. Therefore, it is likely
that all of the residues required for iron sulfur cluster ligation are present in the ZBD. We speculate
that the reduced stability may be due to increased solvent exposure of the iron sulfur cluster in the
absence of the larger RPD domain, which may indicate the iron sulfur cluster is located at the
interface of the ZBD and RPD.

8

Figure 3. The NTD has a modular architecture consisting of a ZBD and RPD, similar to T7
gp4. SDS-PAGE (right) and anti-his immunoblot (left) analysis of 36 kDa N24-A333 construct
after limited trypsin digestion.
Partial digestion patterns suggest the NTD has a modular architecture consisting of a ZBD
and RPD, similar to T7 gp4.
Limited trypsin proteolysis of the NTD (36 kDa) construct produced two stable fragments
of 10 kDa and 26 kDa (Figure 3 above). Immunoblot analysis with the anti-his antibody (Ab)
stained the 10 kDa species but not the 26 kDa species, indicating the 10 kDa fragment contains the
N-terminal His-tag. This domain architecture is consistent with an N-terminal ZBD tethered
flexibly to a larger RPD (26 kDa), as is the case in the N-terminal half of T7 primase-helicase. We
note that the 10 kDa proteolytic fragment is smaller than the 12 kDa ZBD construct. We also
performed the limited trypsin proteolysis in the presence of DNA and RNA substrates specify
which but no change in the digestion profile was observed, indicating no conformational changes
that would protect or influence cleavage occur in the presence of these substrates.
Electronic absorption spectra of purified Dm helicase fractions
Electronic absorption spectrometry measures the absorption of light in the visible to
ultraviolet frequency spectrum of a sample. The absorption peaks pertain to allowed electronic
transitions between energy states, and can give useful information about the electronic structure of
9

the molecule, particularly conjugated π systems and transition metal orbitals. Proteins absorb
radiation at about 280 nm, caused by π to π* transitions of aromatic amino acids, and this method
has been used to quantify the amount of protein in a given sample. UV-visible spectroscopy is not
a conclusive technique, as a single peak could be attributed to many different compounds with the
same energy difference between their HOMO (highest occupied molecular orbital) and LUMO
(lowest unoccupied molecular orbital). Nevertheless, this technique can be useful as supporting
evidence for a particular hypothesis.
The UV-visible spectra collected for the NTD construct show a peak at 280 nm
corresponding to aromatic amino acids in the protein, a shoulder at about 333 nm, and a peak at
about 425 nm (Figure 4 on page 11). For the purposes of this thesis, I have analyzed these UVvisible spectra extensively and have made specific comparisons to previous literature. However,
I have recognized that for the study of proteins with iron sulfur clusters in the biochemical literature
that this type of analysis is not generally used and does not yield conclusions to distinguish 2Fe2S, 3Fe-4S, and 4Fe-4S cluster types. Typically this method is used for initial identification of
iron sulfur clusters, and more specialized spectroscopic methods such as electronic paramagnetic
resonance are used for further characterization.

10

Figure 4. UV-visible electronic absorption spectrum of NTD. Peaks at 422 nm, 464 nm, and
590 nm suggest the presence of an iron-sulfur cluster by comparison with published examples of
UV-visible electronic absorption spectra of protein bound 2Fe-2S and 4Fe-4S clusters.

Figure 5 Comparison of UV-visible spectrum of NTD vs ZBD purified fractions. Pink trace
represents NTD spectrum, both red and green are ZBD fractions after Ni-NTA fractionation.

11

Optical features of as isolated NTD helicase samples and comparison to ZBD spectra
Figure 4 above shows the UV-visible spectrum of an NTD sample (1 mg/ml, 28.5 μM).
Experimental molar extinction coefficients ε (cm-1 M-1) can be calculated for each peak by the
equation A = ε×b×c, where A is the absorbance (y axis), b is the cuvette path length (0.5 cm in this
study), and c is the molarity (mol/ L) of the analyte (7). The 280 nm peak arising from π→π*
transitions of aromatic residues has an experimental extinction coefficient (ε = 30000 cm-1M-1)
that is in excellent agreement with the theoretical value (29910 cm-1 M-1) as predicted from amino
acid sequence (see appendix on page 45). The shoulder centered on 330 nm (ε = 10500 cm-1M-1)
and the broad peak at 420 nm (ε = 7000 cm-1 M-1) correspond to ligand-to-metal charge transfer
bands (opposite cysteine S → Fe) characteristic of iron sulfur clusters (8). Less noticeable but
apparent from the first derivative plot is a broad shoulder around 455 nm (5600 cm-1 M-1) that
also arises from ligand-to-metal charge transfer transitions, and is most commonly observed in
2Fe-2S clusters (9). Spectra obtained for the ZBD were not as reliable as the NTD spectra because
of low concentration and scattering due to aggregation in the samples (Figure 5 on page 11). In
these spectra the peak at 420 nm was less intense, but the peak at 330 nm was still apparent.
Extinction coefficients for these peaks were always very low (ε = 500 – 1000 cm-1 M-1) as
compared to those for the NTD.
Spectral features under reducing conditions
UV-visible electronic absorption spectra of peak fractions of the NTD construct from a
preparation in which the last step used was velocity sedimentation had a spectrum characteristic
of iron-sulfur cluster binding proteins with peaks at 330 and 420 nm (Figure 4 on page 11).
12

Bleaching of the 420 nm peak was observed immediately upon reduction with dithionite (Figure
6 below), and slowly over time in the presence of DTT (Figure 7 on page 14). Both reactions
exhibit unchanged first derivative spectra at wavelengths that are not also absorbed by the reducing
agents (>350 nm for DTT and >400 nm for dithionite) (10). Decreased absorbance associated with
no change in spectral slope characteristics can be interpreted for reduction reactions as either
cluster destruction/ loss or change in electronic structure that influences the capability of this
electronic transition. This reaction to reducing agents is typical of iron sulfur proteins, and could
be caused by destabilization of the cluster leading to cluster loss, especially during
chromatographic methods used in purification (11, 12). The single electron reduction of the cluster
could also disrupt significantly or change its electronic structure, which will impact the energy and
intensity of observed transitions (9, 13, 14, 15, 16). Reduction of the cluster will decrease the
electropositive character of iron atom(s) that could disfavor the ligand-to-metal charge transfer
responsible for the 420 nm absorption peak.

Figure 6. UV-Visible spectrum of the NTD construct after treatment with a strong
reducing agent. An immediate bleaching effect of dithionite on as isolated NTD Dm helicase
sample (blue) after addition of 8 fold excess dithionite (red). Change in volume was negligible
(2 µL in 500 µL) compared to the observed change in absorption at 420 nm.
13

Figure 7. Effect of DTT on the NTD Dm helicase sample after 36 hours on ice (red). As
isolated NTD (blue) is less intense in 300-325nm range. Concentration of as isolated NTD Dm
helicase is 28.5 μM. The oxidation of DTT to the cyclic configuration is accompanied by
increased absorption below 270-340 nm.
Spectral features of in vitro reconstituted samples
A slow increase in the absorbance at 420 nm was observed after the addition of
stoichiometric quantities of ferrous iron (Fe2+) and sodium sulfide (S2-) to a NTD sample in the
presence of 10 mM DTT (Figure 8 on page 15) (17, 18). The observed increase in absorbance at
420 nm can be due to the assembly of protein bound iron sulfur clusters present in the sample.
These results suggest that in vitro reconstitution has occurred and are supported by the observed
2-fold increase of absorbance at 420 nm. Additionally, the derivative plots are essentially the same
showing that the same type of cluster is being formed during reconstitution, and other types of
clusters are not forming as they would most likely cause noticeable differences in the derivative
spectra (19). The intense absorbance in the UV range is due to DTT; when DTT performs
reduction it will cyclize and have increased absorbance (20, 21).
14

Figure 8. UV-Visible analysis of in vitro reconstitution of iron sulfur clusters. The bottom
most (pink) trace represents 1:1 ferrous iron and sulfide: helicase immediately after addition of
NTD helicase (28.5 μM). The second trace (green) is after 1 hour, the third trace (red) is after
addition of 2:1 ferrous iron and sulfide: helicase, and the top trace (blue) is after 12 hours.
Stability of iron-sulfur cluster in the presence of hydrogen peroxide
Addition of hydrogen peroxide up to 100 fold in excess of helicase resulted in no change
in the 280 nm/ 420 nm normalized absorbance ratio (Figure 9 on page 16). These results indicate
that the NTD helicase can retain its iron sulfur cofactor in the presence of mild and high
concentrations of reactive oxygen species (ROS), and this property could have important
consequences in vivo (21, 22).

15

Figure 9. The Fe-S cluster is resistant to oxidizing conditions. Normalized absorbance plot
for hydrogen peroxide titration of NTD Dm helicase. Blue: 7 μM as isolated NTD Dm helicase,
Red: 10:1 H2O2: helicase, Green: 100:1 H2O2: helicase.
Comparison to literature examples of iron sulfur cluster proteins
A major goal of this spectroscopic investigation was to gain insight into the type of iron
sulfur cluster present in the NTD helicase. Table 1 on page 17 presents a UV-visible spectral
comparison of published compounds containing either 2Fe-2S or 4Fe-4S clusters, or both.
Although the UV-visible spectra of both of these cluster types are very similar, some subtle
differences can be discerned. In the 400-500 nm region, 4Fe-4S clusters tend to have a single
maximum at a slightly lower wavelength than 2Fe-2S clusters, which commonly have additional
(lower intensity) absorption peaks around 450 and 550 nm. Reducing agents are known to
destabilize and bleach both types of clusters (23, 24, 25). However, 4Fe-4S clusters are commonly
very unstable toward reactive oxygen species such as H2O2 and O2, often degrading completely

16

or into a stable 2Fe-2S cluster (21, 22, 23, 26, 27). Although these trends have exceptions, they
contribute evidence towards the NTD helicase containing a 2Fe-2S cluster (8, 28, 29, 30, 31).
Table 1. A comparison of UV-visible electronic absorption spectra of 2Fe-2S and 4Fe-4S
clusters from published studies. The UV-visible spectra collected for the NTD construct (on
page 11) show a peak at 280 nm corresponding to aromatic amino acids in the protein, a shoulder
at about 333 nm, and a peak at about 425 nm.
PUBLISHED FE-S
PROTEIN

2FE-2S SPECTRAL PEAKS
(NM)

4FE-4S SPECTRAL PEAKS
(NM)

SYNTHETIC (32)

325, 420, 450

RUMA (33)

444 (oxidized)

390 (native)

FNR (34)

330, 423, 550

315, 409

FERREDOXIN (9)

414, 455, 592

SOXR (12)

330, 414, 462, 548

XPD (35)

405

ADDAB (36)

325, 410

DING (37)

403

RAD3 (38)

400

PRIMASE (39)

315, 400

Chemical analysis of colored fractions by Graphite Furnace Atomic Absorption
Spectroscopy
Graphite furnace atomic absorption (GFAA) spectroscopy is a commonly-used technique
used to detect very small concentrations (ppb) of transition metals in solution. This method is
known for its extreme sensitivity, and therefore smaller amounts of sample can be used for
analysis. Although the instrument can be time-consuming to troubleshoot to obtain correct
atomization conditions, the amount of precious protein that can be saved by choosing this method
mitigates the difficulties associated with the procedure. The instrument operates by injecting a
small (10-40 μL) volume of sample into a graphite coated cuvette that can be heated to high
17

temperatures (3000° C). Once the element of interest reaches its atomization temperature, it will
achieve swiftly the gas phase, where its characteristic wavelength can be measured for absorption
using monochromatic light and a detector. A linear absorption versus concentration plot is
obtained by use of standards, and the extrapolated best fit function can be used for sample analysis.
GFAA spectroscopy was applied to a purified, concentrated sample of the NTD and ZBD
constructs to determine the iron content per protein molecule, as described below.

Iron content
The iron content was determined for numerous NTD and ZBD purifications by GFAA
spectroscopy and horse heart cytochrome c was used as a standard (cyt c contain 1Fe atom per
protein molecule). For NTD purifications, the iron content varied from 1.3-2.0 Fe atoms per
protein molecule, and from 0.4-0.9 Fe atoms per protein molecule for ZBD purifications. An
example of such an analysis of an NTD and ZBD preparations is shown in Table 2 on page 19,
with uncertainty values that were typical for this analysis.

Zinc content
The NTD and ZBD helicase samples were also tested for zinc content, and the determined
zinc content was not statistically different from the background (data not shown). The standard
cytochrome c sample was analyzed by the same method and the result was the same (as expected),
supporting the reproducibility of the method used to obtain this negative result. The lack of zinc
suggests that the ZBD does not bind zinc like the ZBD of T7 gp4.

18

Table 2. Calculated ratios of Fe: protein for the helicase and cyt c samples from graphite
furnace atomic absorption data. Sample concentration was determined by the absorbance at
280 nm for purified helicase variants and by Bradford for the cyt c sample. All samples were
measured in triplicate. Error bars were calculated by propagated error analysis and all
measurements, dilutions, etc were assigned an uncertainty that was appropriate for the
measurement, instrument, pipette, etc. (ex: my P20 pipette was calibrated and found to have a
standard deviation of 0.6 µL, so any measurement or transfer with this pipette, was given an
uncertainty equivalent to the standard deviation (ex: 9.0 ± 0.6 µL).
Sample

Sample
concentration (μM)

Calculated Fe
concentration (μM)

mols of Fe per
mol of protein

NTD Dm helicase

3.6 ± 0.1

4.7 ± 0.2

1.3 ± 0.1

Horse heart
cytochrome c

6.9 ± 0.4

5.8 ± 0.2

0.85 ± 0.06

ZBD Dm helicase

9.6 ± 0.8

9.2 ± 0.3

0.95 ± 0.23

Sulfide determination by methylene blue colorimetric method
Sulfide content was determined by a colorimetric method (see example in appendix on
page 49) that monitored methylene blue formation via the reaction of sulfide with N, N-dimethylp-phenylenediamine (DMPD) (40). Purified fractions of the NTD were determined to contain 2.2
± 0.2 sulfide atoms per protein. Together, the iron and sulfide content of the NTD samples support
the presence of an iron-sulfur cluster. If the iron sulfur cluster is of the 2Fe-2S type, the iron
content indicates that approximately 25% of the protein is without a 2Fe-2S cluster and may be a
result of degradation over the course of the purification, which is common for iron sulfur cluster
proteins. We cannot rule out the possibility that the iron sulfur cluster is the 4Fe-4S type and
>65% of the total protein is apo-protein.

Electron paramagnetic resonance studies
EPR, is the most common method reported in the literature to identify the iron-sulfur
cluster type. The spectral characteristics of 2Fe-2S, 3Fe-4S, and 4Fe-4S have been well
19

documented in each possible redox state and can be used for comparison. Below is a general
introduction to the method and background of EPR.
An electron has an intrinsic spin that imparts an angular spin momentum (ms) of ± 1/2 h,
where h is Planck’s constant (6.626 × 10-34 J × s). According to Pauli’s exclusion property, two
electrons cannot have the same spin quantum number (ms) if they inhabit the same orbital. This
property gives rise to paramagnetic and diamagnetic molecules, where the latter signifies that all
electrons are paired, whereas the former defines a molecule with unpaired electrons. A molecule
with unpaired electron(s) imparts an angular momentum vector to the molecule, and this property
can be influenced by an applied magnetic field. Therefore, when paramagnetic molecules are
subjected to a magnetic field, their electrons spin momentum can either align with the field or
oppose it, creating a lower and higher energy state, respectively. The difference in energy of the
two states can be probed by EPR. This is done by applying a fixed microwave radiation frequency
to the sample (9.46 GHz in our trials) and varying the magnetic field from roughly 400 to 4000
gauss. The equation that describes this technique is ∆E = gβH , where E is the energy difference
between states, H is the applied magnetic field, B is the bohr magneton (9.274 × 10-21 erg), and g
is a variable pertaining to the environment of the unpaired electron (similar to the shielding
constant in NMR). Absorption peaks on EPR spectra can be used to calculate the g value that
varies greater or less than 2.0023193 (value for a free electron), and is caused by the interaction of
the spin momentum of the electron with the orbital momentum of the molecule. Therefore, the g
value for a given peak will be characteristic for that molecule based on its orbital structure and
symmetry, and can provide insight to its electronic structure.

20

Instrumental interference obscured high field signals
A copper impurity in the low temperature device prevented analysis of the high field g = 2
region of the NTD samples (see appendix on page 47). Signals in this region have been well
characterized for iron-sulfur clusters and can be an essential tool in determining the type(s) of
clusters present in a sample. Repeating EPR analysis correctly should be a priority, as the g = 2
region can reliably identify different cluster types and even variable oxidation states (28, 41, 42,
43, 31, 44, 45).
Low field signals may indicate cluster degradation upon reduction
The spectra shown in Figures 10-12 below were performed by a student prior to my time
in the lab, but the analyses I present are my own. A major difference in the purification of NTD
construct by the previous student was that all buffers included EDTA. Figure 10 on page 22
presents the low field EPR spectra of the NTD, which consists of two sharp peaks at g = 4.3 and g
= 3.5. The g = 4.3 peak is characteristic of rhombic high spin ferric iron present in solution
(possibly chelated by EDTA) or bound at the iron sulfur site by itself due to loss of the other iron
from the cluster (46, 47, 48, 49, 41, 50, 51). Upon addition of excess dithionite, a broad low field
feature appears from g ~ 7 to g ~ 4, with the sharp g = 4.3 signal still evident (Figure 11 on page
22). The new broad feature can be attributed to cluster degradation caused by reduction leading
to the release of ferric iron which can be chelated by EDTA and sulfhydryl ligands (DTT or
dithionite) (50, 52). A similar but less intense spectrum is observed upon addition of the oxidizing
agent potassium ferricyanide (Figure 12 on page 23). In this case it is either the oxidation of
aqueous ferrous iron or the abstraction of iron from iron sulfur clusters that allows for formation
of the EDTA/ DTT iron complexes.

21

Figure 10. Low field EPR spectrum of as isolated NTD Dm helicase. Performed by a student
prior to my time in the lab, sample contains EDTA. High field signals near g = 2 are obscured
by a known copper contaminant in the instrument (see appendix on page 47).

Figure 11. Low field EPR spectrum of NTD Dm helicase after the addition of dithionite.
Performed by a student prior to my time in the lab, sample contains EDTA. High field signals
near g = 2 are obscured (see appendix on page 47).

22

Figure 12. Low field EPR spectrum of NTD Dm helicase after the addition of ferricyanide.
Performed by a student prior to my time in the lab, sample contains EDTA. High field signals
near g = 2 are obscured (see appendix on page 47).
Forster resonance energy transfer (FRET) experiments
To examine a possible function of the NTD in DNA binding, I attempted to perform
fluorescence anisotropy (FA) binding assays with a single-stranded oligonucleotide fluoresceinlabeled (ssDNA-fl) by the NTD. When FA was attempted, quenching of 80% of the fluorescence
intensity occurred, which is the opposite effect of what is expected in an FA analysis. I then
investigated if the Fe-S cluster was within range to mediate Forster resonance energy transfer by
acting as the acceptor to quench the incident fluorescence. Indeed, as shown in Figure 13 below,
titration of increasing amounts of NTD construct increased the quenching of fluorescence. This
quenching was absent when the oligonucleotide was not attached to the fluorescein (by using free
fluorescein).

23

Figure 13. Quenching of fluorescence in the presence of NTD construct. Excitation of
Fluorescein (red)-labeled ssDNA (ssDNA-fl) in the absence of NTD exhibited normal
fluorescence and addition of purified NTD construct (blue) resulted in quenching of
fluorescence. Iron sulfur clusters (brown star) can act as acceptors in FRET and result in
quenching.
Adding increasing concentrations of salt to preformed NTD - ssDNA-fl complexes was
observed to inhibit the quenching, and restore fluorescence intensity (Figure 14 on page 25).
Independent experiments were performed using NaCl and MgCl2 to show that salt inhibition of
quenching was not salt-specific and depended only on total ionic strength (Figure 14); addition of
150 mM NaCl or 70 mM MgCl2 resulted in complete interruption of the fluorescence quenching.
Quenching could also be eliminated or reduced by addition of nucleic acids when added to the
preformed NTD - ssDNA-fl as competitive substrates (Figure 15 on page 26) A polydT substrate
(single stranded DNA, never more than 1-2 kDa) was observed to have a coefficient of inhibition
of ki = 100 nM. This ki observed is similar to the binding affinity determined for the ssDNA-fl
substrate, and suggests that the NTD may bind all ssDNA substrates with the weak affinity of
24

approximately 100 nM. Equilibrium of fluorescence intensity was achieved instantaneously after
aliquots of salt were added, whereas it required up to 10 min for complete equilibration after DNA
solutions were added. Overall, the FRET analyses show that the NTD construct has a weak affinity
for DNA substrate and may contribute a functional role.

1-(F/Fmax) Fraction of quenching by Fe-S cluster

0.8
NaCl
MgCl2

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0

20

40

60

80

100

120

140

Conc (mM)
Figure 14. Increasing ionic strength inhibits the quenching observed by the FRET assay.
Salt ions can compete with the ionic interactions at the interface of DNA – protein complexes
and increase the rate of dissociation and inhibit re-binding. Addition of salt (NaCl and
MgCl2).to a solution containing 10 nM fl-ssDNA/ 5 uM NTD Dm helicase caused an immediate
increase in fluorescent intensity. By inhibiting the binding of fl-ssDNA to the NTD the
quenching of fluorescence by FRET is prevented because the Fe-S cluster is no longer in close
proximity to the fluorescein. Complete inhibition of FRET quenching is observed at 150 mM
NaCl and 70 mM MgCl2.
25

100
polydT
polyrA
polyrA-dT

% inhibition

80

60

40

20

0
0.1

1

10

100

1000

Conc (uM)

Figure 15. FRET assay shows that quenching can be inhibited competitively by unlabeled
ssDNA. Semi-log plots of the FRET competitive binding assay comparing the polydT substrate
with ssRNA substrate, polyrA. Concentrations of DNA and RNA substrates are by nucleotide
concentration because the length of individual molecules are not constant.
Proposal and design for mutagenesis of the NTD construct
The spectroscopic and chemical analyses of the purified NTD construct support the
presence of an iron sulfur cluster but the residues responsible for ligation of the cluster remain to
be determined. The majority iron sulfur clusters are coordinated by 4 cysteine residues and applies
to both the 2Fe-2S and 4Fe-4S types. Histidine can substitute for one or more of the cysteine
ligands but examples of alternative residues participating in iron sulfur coordination are much less
26

common. Of the 10 cysteines present in the NTD construct, 9 are conserved among insects and are
highlighted in Figure 16 on page 28. Due to the diversity of iron sulfur cluster motifs that have
been reported in the literature, all 9 cysteine residues were selected for mutagenesis and will be
substituted with an alanine residue. 6 of the 9 selected cysteine residues exist as CXXC motifs
(where X represents any amino acid) so each of the 3 cysteine pairs (C68/C71, C102/C105, and
C245/C248) will be produced as double alanine substitution variants (CXXC→AXXA) because
this CXXC motif is a very common iron sulfur motif in which both cysteines participate in cluster
coordination. The 3 remaining cysteine residues (C63, C260, and C297) will be produced as single
alanine substitution variants for a grand total of 6 alanine substitution variants of the NTD
construct.
The variants most likely to bind the iron sulfur cluster are C68A/C71A and C102A/C105A
because these are the cysteines that align with the 4 cysteine ligands of the ZBD subdomain that
coordinate a zinc ion in T7 gp4 (see figure 2 on page 5). These 4 cysteines of T7 gp4 are also
CXXC motifs and it will be very interesting to examine how four strictly conserved cysteines that
have been preserved in identical motifs are able to bind zinc vs Fe-S clusters. Perhaps the analysis
will show that the cysteines haven’t changed and it’s the surrounding residues that are important
for binding specific metal ions. These residues are present in the ZBD construct of Dm helicase
and although the construct showed limited stability during purification, the ZBD polypeptide coeluted with a light-brown color and suggests the possibility that the ZBD construct can bind the
iron sulfur cluster because all the cysteine residues involved in ligation are located in the ZBD.

27

Figure 16. Alignment of fly, frog and human mtDNA helicases for mutagenesis design and
selection of conserved cysteine residues. Cysteines highlighted in green will be produced as
single alanine substitutions in the NTD construct and cysteines highlighted in cyan will be
produced as double alanine substitutions for each pair of CXXC motifs. Red blocks (α-helical)
and blue arrows (β-sheet) below the sequence illustrates the predicted secondary structure.

28

Experimental approach and possible outcomes
Purification of each variant can be performed via the same steps and conditions used to
purify the wild type NTD construct. Comparing the elution patterns, SDS-PAGE gel mobility, etc
of the variants with wild type NTD purifications can serve as a method to evaluate stability and
physical properties of the variants relative to wild type NTD. In addition, UV-Visible spectra
should be acquired for all variants after each fractionation step in purification to evaluate the
electronic state of the iron sulfur cluster. Variants that exhibit altered spectral characteristics from
the wild type NTD spectra will provide evidence for which cysteines are involved in cluster
coordination. The optimal outcome for this mutagenesis analysis would be for a variant to purify
as a colorless solution and maintain stability. The absence of an iron sulfur cluster eliminates the
possibility for analysis by the FRET binding assay but would allow the variant to be analyzed by
fluorescent anisotropy for DNA binding. The same fluorescently-labeled oligo can be used for
both FRET and fluorescent anisotropy allowing the two methods to be compared accurately to
evaluate the effect of the iron sulfur cluster on DNA binding affinity.

29

DISCUSSION
Previous work by our lab and others has focused mainly on the C-terminal helicase domain
and the linker region of the mtDNA helicase, and has shown that the DNA helicase, NTPase, and
oligomerization characteristics of the mtDNA helicase are very similar to those of the T7 primasehelicase. The N-terminal domain of metazoan mtDNA helicase has lost critical residues required
for the primase activity, which has been studied extensively in T7 primase-helicase. Despite losing
a crucial function, the N-terminal domain retains strong conservation throughout metazoa, and
several disease mutations have been identified that affect the NTD, arguing that the NTD has
retained an old or gained a new role, that is vital for mtDNA metabolism.
Our analysis of the NTD of Drosophila MtDNA helicase presents the first evidence that a
metazoan has retained the modular architecture inherent to bacterial DnaG-like and T-odd phage
primases. In addition, the highly conserved Motif I that binds zinc in DnaG-like primases has been
transformed into an Fe-S cluster-binding motif, presenting the first example of such a change, to
our knowledge. Furthermore, all four cysteines of Motif I, which have been shown to ligate zinc,
are strictly conserved in fly, suggesting the same side chains act as the ligands for Fe-S cluster
binding. The NTD construct bound Fe specifically and showed no presence of Zn, highlighting
the powerful influence that the local environment of the binding site has in discrimination of ligand
binding.
The characterization of the iron-sulfur is only preliminary despite our attempts to perform
EPR; although we obtained electron paramagnetic resonance spectra of the purified NTD construct
in the presence and absence of dithiothreitol, a strong reducing agent. In the absence of reducing
agent we observed a sharp signal at g = 4.3, which is characteristic of ferric iron (Fe+3), and we
observed no change in this signal upon the addition of reducing agent or oxidizing agent

30

(ferricyanide). EPR can be a powerful tool for the identification of iron-sulfur clusters, but only
if one is able to reduce the cluster quickly and stably for the duration of the EPR scan. Iron-sulfur
clusters are typically EPR-silent in the oxidized state, but exhibit signal motifs in the g = 2 region
that are unique to the specific type of iron-sulfur moiety that is present, for example, a 2Fe-2S,
3Fe-4S, or 4Fe-4S. Unfortunately, the only peak we observed in the g = 2 region (see Figure 19
on page 47) was due to a copper contaminant in the instrument itself, and we observed no change
in this background signal throughout the analysis. The g = 4.3 signal indicates the presence of free
or partially-bound ferric iron in the NTD samples, and could be caused by Fe-S cluster degradation
during purification or sample preparation.
A ratio of 1.3-2.0 iron atoms per protein determined by GFAA is consistent with a 2Fe-2S
cluster but does not exclude the possibility of 3Fe-4S or 4Fe-4S clusters. This low amount could
be caused by substantial cluster loss during purification due to the use of reducing agents, which
have been shown to cause cluster degradation by EPR and UV-visible methods. The UV-visible
spectra are typical of 2Fe-2S and 4Fe-4S proteins, as is the observed cluster stability in the
presence of reactive oxygen species, and in the presence of various reducing and oxidizing agents
(21, 22, 23, 26, 27).

31

Conclusions
Overall, the studies presented here for the ZBD and NTD constructs of the Drosophila
melanogaster mtDNA helicase only yield limited conclusions. The UV-visible spectra, iron
content, and sulfide analyses indicate that the NTD construct contains an iron-sulfur cluster, but
the cluster type remains inconclusive.

The EPR spectra do not contribute any significant

information other than that the iron sulfur cluster may be unstable in our purified fractions. The
FRET binding experiments can be used to measure a weak affinity of the NTD construct for DNA
substrates, and this interaction may be functional. Further studies must be performed to add upon
this preliminary data.

32

METHODS
Bacterial cell growth and protein overexpression
E. coli cells containing IPTG-inducible plasmid with cDNA encoding Drosophila
melanogaster mitochondrial helicase residues N24-A333 with N-terminal 6x His tag were grown,
induced, and harvested via centrifugation to produce pellets that were frozen in liquid nitrogen and
stored at -80º C until purification.
Cell lysis and preparation of soluble fraction (FrI)
Pellets were thawed on ice and resuspended in Tris-sucrose buffer (50 mM Tris-HCl pH
7.5, 10% (w/v) sucrose, 5 mM β-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, 10 mM
sodium metabisulfite, 2 µg/mL leupeptin). 5x Lysis buffer (1.25M NaCl, 7.5% n-dodecyl-β-Dmaltoside, 5 mM β-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, 10 mM sodium
metabisulfite, 2 µg/mL leupeptin) was added in a volume 1/5 of the resuspended solution (Final
concentration ~0.25 M NaCl, 1.5% n-dodecyl-β-D-maltoside). Solution was frozen with liquid
nitrogen and thawed on ice. Solution was centrifuged at 12 krpm for 50 minutes at 4º C in a SS34 rotor. Supernatant was collected and pellet discarded. Golden-brown supernatant (FrI) is
soluble fraction and total protein was determined by colorimetric Bradford method.
Ni-NTA metal affinity chromatography (FrII)
A column of Ni-NTA agarose resin was prepared in a ratio of 1 mL resin/ 50 mg total
protein. The column was equilibrated by flowing 10 column volumes (CV) of equilibration buffer
(35 mM Tris-HCl pH 7.5, 500 mM KCl, 10% (v/v) glycerol, 15 mM imidazole, 5 mM βmercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, 10 mM sodium metabisulfite, 2 µg/mL
leupeptin) through the resin. FrI was loaded onto the column at 2 CV/hr. Column was washed
with 10 CV of wash buffer (35 mM Tris-HCl pH 7.5, 500 mM KCl, 10% (v/v) glycerol, 20 mM
imidazole, 5 mM β-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, 10 mM sodium
33

metabisulfite, 2 µg/mL leupeptin) and 1/4 CV fractions were collected. Column was washed with
4 CV elution buffer 1 (35 mM Tris-HCl pH 7.5, 500 mM KCl, 10% (v/v) glycerol, 200 mM
imidazole, 5 mM β-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride, 10 mM sodium
metabisulfite, 2 µg/mL leupeptin) to elute target protein and 1/8 CV fractions were collected.
Column was washed with 3 CV elution buffer 2 (35 mM Tris-HCl pH 7.5, 500 mM KCl, 10%
(v/v) glycerol, 500 mM imidazole, 5 mM β-mercaptoethanol, 1 mM phenylmethylsulfonyl
fluoride, 10 mM sodium metabisulfite, 2 µg/mL leupeptin) to elute any residual protein bound to
column and 1/4 CV fractions were collected. Collected fractions were analyzed via SDS-PAGE
(12% polyacrylamide gels) and fractions were pooled according to target protein content and purity
(FrII). Total protein content of FrII was determined to be 10 mg/ mL by colorimetric Bradford
method and SDS-PAGE analysis estimated a purity of 85%.
Glycerol gradient sedimentation (FrIII)
12-30% glycerol gradients were prepared (35 mM Tris-HCl pH 7.5, 250 mM NaCl, 5 mM
β-mercaptoethanol). FrII was loaded onto gradients and centrifuged at 38 krpm for 72 hrs at 4º C
in a SW-40 rotor. Gradients were fractionated, analyzed by SDS-PAGE (12% polyacrylamide
gels), and fractions were pooled according to purity (FrIII). Total protein content of FrIII was
determined by colorimetric Bradford method and SDS-PAGE analysis revealed no contaminant
bands suggesting a purity of >95%.
Graphite furnace atomic absorption analysis of iron content of NTD (or ZBD) purified
fractions
A Hitachi Z-9000 GFAA spectrometer was used for iron determination courtesy of Dr
Kathryn Severin. Iron standard solutions were prepared from a 1 mM iron (III) nitrate nonahydrate
(crystal, Sigma) solution containing 2% HNO3 (70%, EMD) for each concentration listed in Table
1. Standard solutions additionally contained 3.33% glycerol (100%, J.T. Baker), 0.167 mM EDTA
34

disodium dehydrate (crystal, 100.8%, J.T. Baker), and 11.67 mM Tris-HCl pH 7.5 (Ultrapure,
Research Organics). Digestion of protein samples was achieved by adding 76.0 ± 0.5 μL of the
helicase construct (or 8.1 ± 0.5 μL of cyt c) to 0.5mL HNO3 and 0.5mL H2O2 in a 5mL glass vial.
This solution was heated carefully until boiling with occasional manual stirring to avoid bumping.
Heating was continued until complete evaporation of solution, and the resulting ash was redissolved by adding 200.0 ± 0.5 μL HNO3, and then subsequently diluted to 3.00 ± 0.01 mL with
doubly distilled H2O. Digested samples were sealed with parafilm to avoid iron contamination
from screw-caps during storage. Standard solutions and digested samples were ran in triplicate on
the GFAA spectrometer using a drying temperature of 80-120° C for 30 seconds, ashing
temperature of 630° C for 30 seconds, and an atomization temperature of 2700° C for 10 seconds.
Absorbance values were recorded at a wavelength of 248.3 nm.
Quenching of fluorescence of fluorescein-labeled ssDNA (ssDNA-fl) by NTD via FRET
Fluorescence intensity was measured from 500-550 nm using a Jobin Yvon Spex
Fluorolog-3 spectrofluorimeter with an excitation wavelength of 480 nm. In a 1 mL cuvette, 500
uL of fl-dna solution (10 nM fl-dna oligo/ 50 mM tris-HCl pH 7.5) was added and the fluorescence
spectrum was recorded. A small aliquot (1-2 uL) of concentrated NTD solution was added to this
solution for a total concentration of 10 nM NTD and mixed. Record fluorescence spectrum for
this solution and continue for variable NTD concentrations from 10 nM to 5000 nM by successive
addition of 2 uL aliquots of purified NTD. It was often necessary to wait 2-10 min after mixing
until equilibrium was reached.

35

Calculate dissociation constant Kd for fl-dna binding by NTD
Convert data and plot bound protein (BP) versus free protein (FP). Conversion is done by
assuming 1 binding site per fl-dna for NTD, so BPmax = 10 nM. Fit plot using the following
equation to determine Kd.

𝐵𝑃 =

𝐵𝑃𝑚𝑎𝑥 𝐹𝑃
𝐾𝑑 + 𝐹𝑃

Methylene Blue Photometric Analysis of Sulfide Content in Protein Samples
This general method was used to determine the sulfide content of protein samples suspected
of containing iron sulfur clusters. Add protein solution to a microcentrifuge tube in an amount
equivalent to 1 nmol protein. For standard solutions add an amount of 1 mM Na2S•9H2O in 0.1
M NaOH equivalent to 0.5, 1, 2, 3, and 4 nmol S2-. Dilute to 100 µL with ddH2O. Add 300 µL
1% aqueous solution of zinc (II) acetate dihydrate then add immediately 15 µL 12% aqueous
NaOH. Perform additions to one tube at a time. Let tubes sit for 30 min with inversions. Underlay
suspension with 75 µL DMPD solution (0.1% N,N-dimethyl-p-phenylenediamine (DMPD)•HCl
in 5 M HCl), then add 30 µL FeCl3 solution (23 mM FeCl3 in 1.2 M HCl ) to the bottom layer and
invert the tube immediately to make the solution homogenous. Centrifuge tubes at 2 krpm for 20
minutes. Transfer to cuvette and record visible spectrum from 600-800 nm after at least 30 minutes
have elapsed since the addition of FeCl3. Sulfide reacts with DMPD in the presence of Fe3+ at
acidic conditions (optimal pH = 0.4-0.7) to form methylene blue chromophore which can be

36

detected quantitatively by absorbance at 670 nm (ε = 34500 cm-1 M-1). See appendix on page 49
for example spectrum of standards.

pH = 0.4-0.7
H2S + 2

N

NH2

N

Fe3+

S

N+

N

Inhibition of FRET quenching by salt
Fluorescence intensity was measured from 500-550 nm using a Jobin Yvon Spex
Fluorolog-3 spectrofluorimeter with an excitation wavelength of 480 nm. In a 1 mL cuvette, 500
uL of fl-dna solution (10 nM fl-dna oligo/ 50 mM tris-HCl pH 7.5) was added and the fluorescence
spectrum was recorded.

Concentrated (10 mg/mL) NTD solution was added for a final

concentration of 5 uM to achieve max quenching and fluorescence spectrum was recorded after
mixing and waiting 10 min. A small aliquot (1-2 uL) of concentrated salt solution (NaCl or MgCl2)
was added to this solution for a total concentration of 5-10 mM salt and mixed. Record
fluorescence spectrum for this solution and continue for variable NaCl or MgCl 2 concentrations
(10-1000 mM) by successive addition of aliquots of 3M NaCl or 1M MgCl2.
Competitive FRET binding assay for polynucleic substrates
Fluorescence intensity was measured from 500-550 nm using a Jobin Yvon Spex
Fluorolog-3 spectrofluorimeter with an excitation wavelength of 480 nm. In a 1 mL cuvette, 500
uL of fl-dna solution (10 nM fl-dna oligo/ 50 mM tris-HCl pH 7.5) was added and the fluorescence
spectrum was recorded.

Concentrated (10 mg/mL) NTD solution was added for a final

concentration of 5 uM to achieve max quenching and fluorescence spectrum was recorded after
mixing and waiting 10 min (this spectrum is used for F0 in calculations below) . A small aliquot
37

(1-2 uL) of concentrated substrate solution (polydT, polyrA, polyrA: dT) was added to this solution
for a total concentration of 5 nM-1 uM substrate (determined as nucleotide concentration) and
mixed. Record fluorescence spectrum for this solution and continue for variable substrate
concentrations (10 nM-100 uM determined as nucleotide concentration) by successive addition of
aliquots of concentrated substrate (polydT, polyrA, polyrA:dT).

After the max substrate

concentration is reached, NaCl was added to a final concentration of 150 mM and spectrum was
recorded. Addition of salt is used to determine Fmax, defined as the fluorescent intensity when
NTD- fl-dna binding is completely inhibited.
Calculate the inhibition coefficient Ki for each substrate
Determine IC50 (inhibitor concentration at 50% inhibition) by fitting a function: y = a +
b*log(x), where x = [I] (concentration of inhibitor), y = % inhibition. Once a and b are determined,
use the function to determine x when y = 50. The % inhibition is defined as (F/F*max)*100, where
F = the fluorescent intensity at 522 nm for a given [I], F*max is Fmax – F0, F0 is the fluorescent
intensity at 522 nm when no inhibitor is present and Fmax, is the fluorescent intensity at 522 nm
when NTD-fl-dna binding is completely inhibited. Determine Ki via the following equation,
where Kd =122 nM and [S] = [NTD] = 5 uM. See appendix on page 48 for example of plot and
fit used to calculate the IC50.

𝐾𝑖 =

𝐼𝐶50 𝐾𝑑
𝑆 + 𝐾𝑑

38

EPR Spectroscopy
100-150 uL samples of concentrated NTD solutions (10 mg/ mL) were frozen in liquid
nitrogen prior to analysis. For oxidized samples, 10-fold excess potassium ferricyanide was added
10 min prior to freezing. For reduced samples, 10-fold dithionite was added 10 min prior to
freezing. All spectra were recorded on instruments in the McCracken lab. Baselines were recorded
using sample buffer (35 mM tris-HCl/ 50 mM NaCl/ 15% glycerol). All experiments were
performed at 4 K.
UV-Visible Spectroscopy
All spectra were recorded on an Agilent 8453 UV-Visible spectrophotometer. For all
analyses, a quartz cuvette with a pathlength of 0.5 cm was used and required a minimal volume of
0.5 mL.

Methods for mutagenesis
Growth and preparation of Ca2+ competent cells
Add 1 mL of an overnight culture of bacteria to 20 mL of LB brother in a 125 mL flask.
Grow the cells at 37º C in a water bath with shaking. Monitor OD595 every 0.5 hr and record/plot
growth. When the culture has OD595 = 0.2 place the flask on ice for 3-5 min to chill the bacteria.
Transfer to a chilled Oakridge tube and centrifuge in the SS-34 rotor at 10 krpm for 2 min. Decant
supernatant and resuspend cells in 2.5 mL of ice-cold 0.05 M CaCl2. Leave cells on ice for >15
min. Centrifuge cells in the SS-34 rotor at 10krpm for 2 min and remove supernatant with a pipette
instead of decanting. Resuspend cells in 2.5 mL of ice-cold 0.05 M CaCl2.

39

Transformation of Ca

2+

competent cells

To 5 micro-centrifuge tubes on ice, add 50 µL of TCM. Add DNA- (Plasmid from miniprep) to tubes 1, 2, and 3 in the amount of 1 µL, 2 µL, and 3 µL, respectively, and leave 2 tubes
without DNA for controls by later plating them +/- 25 µg/ mL kanamycin. To each tube add 100
µL of cells. Flick the tube gently with your finger and incubate on ice > 20 min. Incubate tubes for
2 min at 42º C. Add 1 mL LB broth to each tube. If you plan to select for drug resistance incubate
30 min in 37º C (WITH NO DRUG ADDED TO BROTH). Otherwise plate on LB plates. Plate
the transformed cells on LB plates containing drug by spreading or by addition of 2.5 mL of top
agar to the transformed cells after putting your transformed cells into 13 x 100 tubes. Incubate
plates for 1-2 days. If plate for plaques with M13 or derivatives of M13, add indicator bacteria
(0.2-0.3 mL of an overnight culture to the sample and use top-agar for plating). If you are using
one of the M13 mp phages add 10 µL of 0.1 M IPTG, 50 µL 2% X-Gal to top agar.
Preparation of electrocompetent bacteria (XL-1 Blue)
Inoculate a day culture in LB medium with appropriate antibiotics (15 µg/mL tetracycline)
by diluting an overnight culture 200 times (500-1000 mL of culture, 250 mL per 1L flask). Grow
bacteria with shaking at appropriate temperature until OD595 is about 0.35-0.40.

Take

measurement every 30 min for growth curve. At an OD595 of 0.35-0.40 place bacteria on ice for
at least 30 min. Harvest bacteria by spinning in 250 mL bottles in Sorval GSA rotor for 10 min at
8,000 rpm. Resuspend bacterial pellet in original volume of ice-cold sterile ddH2O. Keep on ice
for at least 10 min. Harvest bacteria by spinning as above. Gently remove supernatant. Bacterial
pellet is very fragile at this point. Resuspend the pellet in 1/2 original volume of ice-cold sterile
ddH2O. Spin in 250 mL bottles in Sorval GSA rotor for 10 min at 8,000 rpm. Resuspend the pellet

40

in 5 mL ice-cold sterile 10% (v/v) glycerol and transfer into oackridge tubes (equivalent of 500
mL of original culture per tube). Spin at 8000 rpm for 10 min. Remove supernatant, resuspend the
pellet in 0.6 mL of sterile 10% (v/v) glycerol. Make aliquots of 130 µL and flash freeze in liquid
N2. Store at -70º C.
Mini-preparation of plasmid DNA: pET-28a- Dm mt helicase N24-A333 from BL21(DE3)
cells
To obtain pET-28a-dm mt helicase N24-A333 DNA from BL21(DE3) bacterial cells,
Inoculate two colonies from pET-28a-dm mt helicase N24-A333 BL21(DE3) plate in 2 separate
tubes of 3 mL liquid L-broth w/ 15 µg/ mL kanamycin. Grow cultures overnight at 37º C w/ shaker.
Centrifuge cultures in micro-centrifuge tubes for 1 min at 13000 rpm. Discard supernatant.
Resuspend cells in 300 µL of P1 (50 mM Tris-HCl pH 7.5/ 10 mM EDTA pH 8.0/ 0.1 mg/mL
RNase) by vortex. Make sure all cells are resuspended. Add 300 µL of P2 (200 mM NaOH/ 1%
SDS) and invert tubes ~5 times. Incubate at room temperature for no longer than 5 min. Add 300
µL of P3 (3M potassium acetate pH 5.5), mix by inversion, and incubate 2 min at room
temperature. Centrifuge at 13000 rpm for 10 min. Collect supernatant and transfer to new microcentrifuge tubes. Add 400 µL 100% isopropanol, incubate 5 min, and centrifuge at 13000 rpm for
10 min. Discard supernatant. Add 600 µL 70% ethanol and resuspend pellet. Centrifuge again at
13000 rpm for 5min. Discard supernatant. Let the pellet sit out for 20 min to dry and resuspend in
20 µL of ddH2O.
Transformation of pET-28a-dm mt helicase N24-A333 plasmid into XL-1 Blue cells via
electroporation
Thaw one tube of XL-1 Blue competent cells on ice. Distribute 40 µL into one new tube.
For controls, distribute 40 µL into a separate tube and do not add any plasmid. Add 1 µL of miniprep DNA and mix gently for 2 min. Transfer cells/plasmid to the E. Coli Pulser cuvette, set the
41

voltage of the electroporator to 2.3 V and hit on. Transfer electroporated cells immediately to 1
mL L-broth (WITHOUT DRUGS) and grow them at 37º C for ~1 hr w/ shaker. Plate 20, 100, and
500 µL on separate agar plates containing drug. Plate 500 µL of the control on a drug containing
plate and 500 µL on a plate without drugs. Incubate plates overnight at 37º C.
Mini-preparation of plasmid DNA: pET-28a- Dm mt helicase N24-A333 from XL-1 Blue cells
To obtain methylated pET-28a-dm mt helicase N24-A333 DNA from XL-1 Blue bacterial
cells, Inoculate two colonies from pET-28a-dm mt helicase N24-A333 XL-1 Blue plate in 2
separate tubes of 3 mL liquid L-broth w/ 15 µg/ mL kanamycin. Grow cultures overnight at 37º C
w/ shaker. For both overnight cultures, transfer 1 mL into each of 3 micro-centrifuge tubes and
centrifuge for 1 min at 13000 rpm. Discard supernatant. Resuspend cells in 100 µL of P1 (50 mM
Tris-HCl pH 7.5/ 10 mM EDTA pH 8.0/ 0.1 mg/mL RNase) by vortex. Make sure all cells are
resuspended. Combine the three 100 µL suspensions into 1 tube for both cultures. Add 300 µL of
P2 (200 mM NaOH/ 1% SDS) to both tubes and invert tubes ~5 times. Incubate at room
temperature for no longer than 5 min. Add 300 µL of P3 (3M potassium acetate pH 5.5) to both
tubes, mix by inversion, and incubate 2 min at room temperature. Centrifuge at 13000 rpm for 10
min. Collect supernatant and transfer to new micro-centrifuge tubes. Add 400 µL 100%
isopropanol, incubate 5 min, and centrifuge at 13000 rpm for 10 min. Discard supernatant. Add
600 µL 70% ethanol and resuspend pellet. Centrifuge again at 13000 rpm for 5min. Discard
supernatant. Let the pellet sit out for 20 min to dry and resuspend in 20 µL of ddH2O.
Quick change mutagenesis of dm mtDNA helicase N24-A333
Dilute mini-prep to 10 ng/µL. Prepare master mix (0.2 mM dNTP’s, 1X pfu stratagene
turbotag buffer, 2.5 units/ reaction pfu turbotag pol stratagene). Add 1 µL of each primer and 5
µL plasmid. PCR program consisting of 25 cycles of 95º C for 30 sec, 55º C for 60 sec and 68º C

42

for 12 min. Remove tubes from thermocycler and add 1 µL of DpnI to all tubes except tube 8.
Incubate at 37º C for 1hr. Add 1 µL of each reaction in 40 µL of XL-1 Blue cells. Mix gently for
1-2 min. Transfer cells/reaction to E. coli pulser cuvette, set the voltage of the electroporator to
2.3 V, and hit on. Transfer electroporated cells immediately to 500 µL L-broth (WITHOUT
DRUG) and grow cells at 37º C for ~1 hr w/ shaker. Spin cells down for 1 min at 13000 rpm.
Resuspend them in 200 µL L-broth and plate on agar plates (WITH DRUG). Incubate plates
overnight at 37º C.

43

APPENDIX

44

APPENDIX

Amino acid sequence of the N-terminally 6x His-tagged NTD (N24-P333) construct of Dm
mtDNA helicase. The sequence of the ZBD (N24-P123) construct is underlined and the 10
additional residues of the 6x His tag are in bold format. The ZBD construct also contains the Nterminal 6x His tag.
MGHHHHHHATN24YATQVVSGLEECSLDPKEYVDFKRQLRQLNLPHKDGHTCLQLEC
RLCDRNRQPVTNAQKGTDHGLLAYVNKRTGAFICPNCDVKTSLTSALLSYQLPKP123
VGYKQPLQRQPVYESRFPHLAVVTPEACAALGIKGLKEDQLNAIGAQWEPQQQLLHFK
LRNAAQVEVGEKVLYLGDRREEIFQSSSSSGLLIHGAMNKTKAVLVSNLIDFIVLATQNI
ETHCVVCLPYELKTLPQECLPALERFKELIFWLHYDASHSWDAARAFALKLDERRCLLI
RPTETEPAPHLALRRRLNLRHILAKATPVQHKA333

Table 3. Physical properties based on primary sequence of NTD helicase. Theoretical
physical properties as determined from amino acid sequence of Dm mtDNA helicase N24-A333.
Molecular Weight (Daltons) Isoelectric Point (pI) Extinction Coefficient (cm-1M-1)
36283.8

8.81

29910 (280 nm)

45

Table 4. Protein content and purity of purification steps. Total protein concentration was
determined by A280 analysis and by the colorimetric Bradford method throughout purification.
Purity was estimated by SDS-PAGE analysis of target band versus contaminant bands.
Purification Step

Total Protein
Concentration by
A280 (mg/ mL)

Total Protein
Concentration by
Bradford (mg/ mL)

Predicted
Purity (%)

FrI

Cell Lysis

ND

7

20

FrII

Ni-NTA metal
affinity
chromatography

ND

10

85

FrIII

Glycerol Gradient
Sedimentation

1.1

1.4

>95

Pool
Name

Figure 17. SDS-PAGE analysis of NTD construct throughout the purification. We suspect
an SDS-resistant disulfide-linked dimer is the band at 70 kDa.

46

Figure 18. Gel filtration analysis of NTD construct using a Superdex 75 column. The elution
of a single peak corresponding to a monomer as determined by the A280 measurements shown as
the blue trace.

Figure 19. Instrument interference at g = 2 prevented further characterization of Fe-S
characteristics by EPR. This EPR spectrum was taken with an empty sample tube to show that
the large signal at g = 2 is suspected to be a copper impurity in the instrument and not due to
sample contamination.
47

100
y = m1 + m2 * log(M0)

% inhibition

80

m1
m2
Chisq
R

60

Value
Error
21.578 1.5962
45.327 1.5216
35.683
NA
0.99775
NA

40

20

0
0.1

1

Conc (uM)

10

Figure 20. Example of semi-log plot and fit used to calculate the IC50. (The x axis is
logarithmic).

48

Figure 21. Example UV-Visible spectrum for standards: The reaction is considered successful
if the ratio of absorbance between A670 and A750 is close to 2 as is observed in these spectra.
Blue- .5 nmol sulfide, Red- 1 nmol sulfide, Pink- 2 nmol sulfide, Dark green- 3 nmol sulfide,
Brown- 4 nmol sulfide.

49

BIBLIOGRAPHY

50

BIBLIOGRAPHY

1. Ziebarth, T., Farr C., Kaguni, L. (2007). Modular Architecture of the Hexameric Human
Mitochondrial DNA Helicase. J. Mol. Biol., 367, 1382-1391.
2. Matsushima, Y., Kaguni, L. (2009). Functional importance of the conserved N-terminal
domain of the mitochondrial replicative DNA helicase. Biochimica et Biophysica Acta,
1787, 290–295.
3. S. S. Patel, K. M. Picha. STRUCTURE AND FUNCTION OF HEXAMERIC
HELICASES1. Annual Review of Biochemistry, 69, 651 -697
4. Shutt TE, Gray MW. (2006). Twinkle, the mitochondrial replicative DNA helicase, is
widespread in the eukaryotic radiation and may also be the mitochondrial DNA primase
in most eukaryotes. Journal of Molecular Evolution, 62, 588-599.
5. Longley MJ, Humble MM, Sharief FS, Copeland WC. (2010) Disease variants of the
human mitochondrial DNA helicase encoded by C10orf2 differentially alter protein
stability, nucleotide hydrolysis, and helicase activity. Journal of Biological Chemistry,
285, 29690-702
6. Ilyina TV, Koonin EV. (1992) Conserved sequence motifs in the initiator proteins for
rolling circle DNA replication encoded by diverse replicons from eubacteria, eucaryotes
and archaebacteria. Nucleic Acid Research, 20, 3279-85.
7. Schmid, F. (2001). Biological Macromolecules: UV-Visible Spectrophotometry.
ENCYCLOPEDIA OF LIFE SCIENCES, Macmillan Publishers Ltd, Nature Publishing
Group / www.els.net.
8. Tsibri, J,. Woody, R. (1970). Structural Studies of Iron-Sulfur Proteins. Coordin. Chem.
Rev., 5417-458.
9. Noodle, L., Baerends, E. (1984). Electronic Structure, Magnetic Properties, ESR, and
Optical Spectra for 2-Fe Ferredoxin Models by LCAO-Xα Valence Bond Theory. J. Am.
Chem. Soc., 106, 2316-2327.
10. Iorio, E. (1981). Preparation of Derivatives of Ferrous and Ferric Hemoglobin.
METHODS IN ENZYMOLOGY, VOL. 76.
11. Hidalgo, E., Bollinger, J., Bradley, Terence., Walsh, C., Demple, B. (1995). Binuclear
[2Fe-2S] Clusters in the Escherichia coli SoxR Protein and Role of the Metal Centers in
Transcription. THE JOURNAL OF BIOLOGICAL CHEMISTRY , Vol. 270, No. 36, Issue
of September 8, pp. 20908–20914.
12. Hidalgo, E., Demple, B. (1994). An iron-sulfur center essential for transcriptional
activation by the redox-sensing SoxR protein. The EMBO Journal,13, 138-146.

51

13. Torres, R., Lovell, T., Noodleman, L., Case, D. (2003). Density Functional and
Reduction Potential Calculations of Fe4S4 Clusters. J. AM. CHEM. SOC., 125, 19231936.
14. Mouesca, J., Chen, Jun., Bashford, D., Case, D. (1994). Density Functional PoissonBoltzmann Calculations of Redox Potentials for Iron-Sulfur Clusters. J. Am. Chem. SOC.
116, 11898-11914.
15. Stephens, P. J., Jollie, D. R., Warshel, A. (1996). Protein Control of Redox Potentials of
Iron-Sulfur Proteins. Chem. Rev. 96, 2491-2513.
16. Noodleman, L., Norman, J., Osborne, J., Aizman, A., Case, D(1985). Models for
Ferredoxins: Electronic Structures of Iron-Sulfur Clusters with One, Two, and Four Iron
Atoms. J. Am, Chem. SOC. 107, 3418-3426.
17. Malkin, R., Rabinowitz, J. (1966). The Reconstitution of Clostridial Ferredoxin.
Biochemical and Biophysical Research Communications, 23, No. 6.
18. Zheng, C., Zhang, Y., Liu, Y., Wu, A., Xia, L., Zeng, J., Liu, J., Qiu, G. (2009).
Characterization and Reconstitute of a [Fe4S4] Adenosine 5’-Phosphosulfate Reductase
from Acidithiobacillus ferrooxidans. Curr. Microbiol., 58, 586-592.
19. Rickard, D., Luther III, G. (2007). Chemistry of Iron Sulfides. Chem. Rev., 107, 514562.
20. Johnson, C., Asher, S. (1987). UV Resonance Raman Excitation Profiles of l-Cystine.
Journal of Raman Spectroscopy, 18, 345-349.
21. Ding, H., Demple, B. (1998). Thiol-Mediated Disassembly and Reassembly of [2Fe-2S]
Clusters in the Redox-Regulated Transcription Factor SoxR. Biochemistry,37, 1728017286.
22. Reents, H., Gruner, I., Harmening, U., Bottger, Lars., Layer, G., Heathcote, P.,
Trautwein, A., Jahn, D., Hartig, E. (2006). Bacillus subtilis FNR senses oxygen via a
[4Fe-4S] cluster coordinated by three cysteine residues without change in the oligomeric
state. Molecular Microbiology, 60, 1432-1445.
23. Lillig, C., Berndt, C., Vergnolle, O., Lonn, M., Hudemann, C., Bill, E., Holmgren, A.
(2005). Characterization of human glutaredoxin 2 as iron-sulfur protein: A possible role
as redox sensor. PNAS, 102, 8168-8173.
24. Bych, K., Kersercher, S., Netz, D., Pierik, A., Zwicker, K., Huynen, M., Lill, R., Brandt,
U., Balk, J. (2008). The iron-sulphur protein Ind1 is required for effective complex 1
assembly. The EMBO Journal, 27, 1736-1746.
25. Dailey, H., Finnegan, M., Johnson, M. (1994). Human Ferrochelatase is an Iron-Sulfur
Protein. Biochemistry, 33, 403-407.
26. Singh, A., Guidry, L., Narasimhulu, K., Mai, D., Trombley, J., Redding, K., Giles, G.,
Lancaster, J., Steyn, A. (2007). Mycobacterium tuberculosis WhiB3 responds to O2 and
nitric oxide via its [4Fe-4S] cluster and is essential for nutrient starvation survival. PNAS,
104, 11562-11567.
52

27. Jervis, A., Crack, J., White, G., Artymiuk, P., Cheesman, M., Thomson, A., Brun, N.,
Green, J. (2009). The O2 sensitivity of the transcription factor FNR is controlled by
Ser24 modulating the kinetics of [4Fe-4S] to [2Fe-2S] conversion. PNAS¸106, 46594664.
28. Sweeney, W., Rabinowitz, J. (1980). Proteins Containing 4Fe-4S Clusters: An Overview.
Ann. Rev. Biochem., 49, 139-161.
29. Orme-Johnson, W. H. (1973) Iron-Sulfur Proteins: Structure and Function. Annu. Rev.
Biochem, 42, 159-204.
30. Koay, M., Antonine, M., Girtner, W., Lubitz, W. (2008). Modeling Low-Potential
[Fe4S4] Clusters in Proteins. Chemistry and Biodiversity, 5, 1571.
31. McLean, K., Warman, A., Seward, H., Marshall, K., Girvan, H., Cheesman, M.,
Waterman, M., Munro, A. (2006). Biophysical Characterization of the Sterol
Demethylase P450 from Mycobacterium tuberculosis. Its Cognate Ferredoxin, and their
Interactions. Biochemistry, 45, 8427-8443.
32. Kennedy, M., Kent, T., Emptage, M., Merkle, H., Beinert, H., Munck, E. (1984).
Evidence for the Formation of a Linear [3Fe-4S] Cluster in Partially Unfolded Aconitase.
The Journal of Biological Chemistry, 259, 14463-14471.
33. Agarwalla, S., Stroud, R., Gaffney, B. (2004). Redox Reactions of the Iron-Sulfur Cluster
in a Ribosomal RNA Methyltransferase, RumA. The Journal of Biological Chemistry,
279, 34123-34129.
34. Khoroshilova, N., Popescu, C., Munck, E., Beinert, H., Kiley, P. (1997). Iron-Sulfur
cluster disassembly in the FNR protein of Escherichia coli by O2: [4Fe-4S] to [2Fe-2S]
conversion with loss of biological activity. Proc. Natl. Acad. Sci.,94, 6087-6092.
35. Rudolf, J., Makrantoni, V., Ingledew, J., Stark, M., White, M. (2006). The DNA Repair
Helicases XPD and FancJ Have Essential Iron-Sulfur Domains. Molecular Cell, 23, 801808.
36. Yeeles, J., Cammack, R., Dillingham, M. (2009). An Iron-Sulfur Cluster is Essential for
the Binding of Broken DNA by AddAB-type Helicase-Nucleases. The Journal of
Biological Chemistry, 284, 7746-7755.
37. Ren, B., Duan, X., Ding, H. (2009). Redox Control of the DNA Damage-inducible
Protein DinG Helicase Activity via its Iron-Sulfur Cluster. The Journal of Biological
Chemistry, 284, 4829-4835.
38. Pugh, R., Honda, M., Leesley, H., Thomas, A., Lin, Y., Nilges, M., Cann, I., Spies, M.
(2008). The Iron-containing Domain is Essential in Rad3 Helicases for Coupling of ATP
Hydrolysis to DNA Translocation and for Targeting the Helicase to the Single-stranded
DNA-Double-Stranded DNA Junction. The Journal of Biological Chemistry, 283, 17321743.
39. Weiner, B., Huang, H., Dattilo, B., Nilges, M., Flanning, E., Chazin, W. (2007). An IronSulfur Cluster in the C-terminal Domain of the p58 Subunit of Human DNA Primase.
The Journal of Biological Chemistry, 282, 33444-33451.
53

40. Rabinowitz, JC. (1978). Analysis of acid-labile sulfide and sulfhydryl groups.
Methods Enzymol., 53, 275–277.
41. Hagen, Wilfred R.(2007) 'Wide zero field interaction distributions in the high-spin EPR
of metalloproteins', Molecular Physics, 105: 15, 2031 — 2039
42. Blumberg, W. E., Peisach, J. (1974). On the interpretation of Electron Paramagnetic
Resonance Spectra of Binuclear Iron-Sulfur Proteins. Achives of Biochemistry and
Biophysics, 162, 502-512.
43. Guigliarelli, B., Bertrand, P., Gayda, J. (1986). Contribution of the fine structure terms to
the g values of the biological Fe III – Fe II clusters. J. Chem. Phys., 85, 1689-1675.
44. Mulder, D. (2005). Fe4S4 clusters: A review of the electronic and geometric properties of
synthetic analogues of Fe4S4 active sites in iron-sulfur proteins. (4Fe4SclusterselectronicandgeometricREVIEW.pdf)
45. Noodleman, L. (1991). Exchange Coupling and Resonance Delocalization in Reduced
[Fe4S4]+ and [Fe4Se4]+ clusters. 1. Basic Theory of Spin-State Energies and EPR and
Hyperfine Properties. Inorganic Chemistry, 30, 246-256.
46. Abdallah, F., Chasteen, N. (2008). Spin concentration measurements of high-spin (g=4.3)
rhombic iron(III) ions in biological samples: theory and application. J. Biol. Inorg.
Chem., 13, 15-24.
47. Marton, A., Sukosd-Rozlosnik, N., Vertes, A., Horvath, I. (1987). The effect of EDTAFe(III) complexes with different chemical structure on the lipid peroxidation in brain
microsomes. Biochemical and Biophysical Research Communications, 145, 211-217.
48. Aisen, P., Aasa, R., Malmstrom, B., Vanngard, T. (1967). Bicarbonate and the Binding of
Iron to Transferrin. The Journal of Biological Chemistry, 242, 2484-2490.
49. Wegner, P., Bever, M., Schunemann, V., Trautwein, A., Schmidt, C., Bonisch, H., Gnida,
M., Meyer-Klaucke, W. (2004). Iron-Sulfur Proteins Investigated by EPR, Mossbauer,
and EXAFS Spectroscopy. Hyperfine Interactions,156/157, 293-298.
50. Philip, C. V., Brooks, D. (1974). Iron(III) Chelate Complexes of Hydrogen Sulfide and
Mercaptans in Aqueous Solution. Inorganic Chemistry, 13, 384-387.
51. Cooper, C., Salerno, J. (1992). Characterization of a Novel g=2.95 EPR Signal from the
Binuclear Center of Mitochondrial Cytochrome c Oxidase. The Journal of Biological
Chemistry, 267, 260-285.
52. Werth, M., Kurtz, Jr, D., Howes, B., Huynh, B. (1989). Observation of S=2 EPR Signals
from Ferrous Iron-Thiolate Complexes. Relevance to Rubredoxin-Type Sites in Proteins.
Inorganic Chemistry, 28, 1357-1361.

54

CHAPTER 2: Clustering of Alpers disease mutations and catalytic defects in biochemical
variants reveal new features of molecular mechanism of the human mitochondrial replicase
Pol γ.
Liliya Euro1, Gregory A. Farnum2, Eino Palin1, Anu Suomalainen1,3 and Laurie S. Kaguni2

1

Research Program of Molecular Neurology, Biomedicum-Helsinki, University of Helsinki,

Haartmaninkatu 8, 00290 Helsinki, Finland, 2Department of Biochemistry and Molecular Biology
and Center for Mitochondrial Science and Medicine, Michigan State University, East Lansing, MI
48824-1319, USA, and 3Department of Neurology, Helsinki University Hospital, Helsinki,
Finland

55

ABSTRACT
Mutations in Pol γ represent a major cause of human mitochondrial diseases, especially
those affecting the nervous system in adults and in children. Recessive mutations in Pol γ represent
nearly half of those reported to date, and they are nearly uniformly distributed along the length of
the POLG1 gene (Human DNA Polymerase gamma Mutation Database); the majority of them are
linked to the most severe form of POLG syndrome, Alpers–Huttenlocher syndrome. In this report,
we assess the structure–function relationships for recessive disease mutations by reviewing
existing biochemical data on site-directed mutagenesis of the human, Drosophila and yeast Pol γs,
and their homologs from the family A DNA polymerase group. We do so in the context of a
molecular model of Pol γ in complex with primer–template DNA, which we have developed based
upon the recently solved crystal structure of the apoenzyme form. We present evidence that
recessive mutations cluster within five distinct functional modules in the catalytic core of Pol γ.
Our results suggest that cluster prediction can be used as a diagnosis-supporting tool to evaluate
the pathogenic role of new Pol γ variants.

56

INTRODUCTION
Mitochondrial DNA polymerase, Pol γ, is the sole known DNA polymerase in animal
mitochondria and is responsible for mitochondrial DNA (mtDNA) replication and repair (1, 2).
The human enzyme is a heterotrimer consisting of a catalytic subunit, Pol γA, and dimer of an
accessory subunit, Pol γB (3). Pol γA is a member of the family A DNA polymerase group to
which bacterial DNA Pol I belongs (1). It comprises three domains: an N-terminal domain
containing 3′→5′ exonuclease (exo) activity, a spacer domain and a C-terminal domain, containing
5′→3′ DNA polymerase (pol) activity in three subdomains termed the palm, fingers and thumb.
Pol γA also bears a 5′-deoxyribose phosphate lyase activity (4), but the location of its active site is
unknown. The accessory subunit, Pol γB, serves as a processivity factor, enhancing the DNAbinding affinity and catalytic activities of Pol γA (5,6). The crystal structure of human Pol γ in its
apoenzyme form was solved recently (PDB code 3IKM) (7).
Mutations in Pol γ represent a major cause of human mitochondrial diseases, especially
those affecting the nervous system in adults and in children. Dominant mutations typically cause
adult-onset myopathies and encephalopathies (8, 9), whereas recessive mutations result in severe
adult or juvenile onset ataxia-epilepsy syndromes (MIRAS, SCA-E, SANDO) (10–12), or
devastating early-childhood Alpers syndrome (Alpers–Huttenlocher syndrome) characterized by
intractable epilepsy, psychomotor retardation and liver failure that leads to early death (13, 14).
On the molecular level, POLG syndromes are accompanied by either tissue-specific mtDNA
depletion, deletions or combination of both (15, 16). To date, more than 145 POLG1 disease
mutations have been identified (Human DNA Polymerase gamma Mutation Database,
http://tools.niehs.nih.gov/polg/), of which more than half are recessive. The functional
consequences of dominant mutations, which affect primarily catalytic residues in the pol domain,
have been explained by effective competition for the DNA substrate by the mutant Pol γ with the
57

wild-type enzyme (7,17,18), and by site-specific stalling of mutant DNA polymerase (18). Studies
that explain effects of individual recessive mutations on protein function are limited (19–22). The
majority of the reported recessive mutations are linked to the most severe form of POLG syndrome,
Alpers–Huttenlocher syndrome. Manifestation of Alpers disease typically requires the presence of
at least two recessive, compound heterozygous mutations, and is accompanied primarily with
mtDNA depletion (16, 23). The detailed mechanisms of pathogenesis and mtDNA depletion are
unknown.
The assignment of a newly identified amino acid substitution as a pathogenic mutation is
generally based on the absence of the variant in the normal population, conservation of the site
among species and segregation in families. These assignments are particularly challenging because
of the extensive variation in the amino acid sequences of Pol γA in the normal population. To date,
data accumulated on recessive disease mutations show that they are almost uniformly distributed
among the three structural domains of the catalytic subunit and in most cases, it is unclear which
property of the enzyme is affected, and how it contributes to mtDNA deletion or depletion.
The present study aims to assess the structure–function relationships for recessive disease
mutations, by reviewing existing biochemical data on site-directed mutagenesis of the human,
Drosophila and yeast Pol γs, and their homologs from the family A DNA polymerase group; we
do so in the context of the recently solved crystal structure of human Pol γ (7), onto which we have
modeled primer–template DNA (ptDNA). We present evidence that recessive mutations cluster
within five distinct functional modules in the catalytic core of Pol γ that we designate as ‘Alpers
Clusters 1–5’. We note that our analysis of the recessive mutations found in compound
heterozygous form in Alpers patients reveals that a severe disease manifestation is typically caused
by a combination of at least two mutations from different clusters. Our results suggest that cluster

58

prediction can be used as a diagnosis-supporting tool to evaluate the pathogenic role of new Pol γ
variants.
Approach to comparative structural analysis
We docked ptDNA into the putative DNA-binding channel of the apoenzyme form of the
human Pol γ structure (PDB code 3IKM) (7) by superposition of the closed ternary complex of T7
Pol bound to ptDNA and dNTP (PDB code 1T8E) (24). The coordinates of the resulting Pol γ:
DNA model are provided in Supplementary Data of the online article. To ensure reliable
positioning for subsequent structural analysis of Alpers recessive mutations, we evaluated three
different alignments of the Pol γ and T7 palm domains that are presented in Figure 29 on page 91.
First, we aligned the palm domain of Pol γA (PDB code 3IKM, chain A, residues 815–910 and
1095–1239) with that of T7 Pol (PDB code 1T8E, chain A, residues 409–487 and 611–704).
Additionally, we aligned the two structures by superposition of their 12β-loop-13β motifs from the
palm subdomain, encompassing residues G1127-E1144 in Pol γA and F646-E663 in the T7
catalytic core [for secondary structure element assignment see (25)]. This part of the central βsheet in the palm subdomain is the most structurally conserved element between DNA polymerases
in the family A DNA polymerase group (25), and it superposed with an RMSD of 1.906 Å for Pol
γA and T7. Alignment showed that the functionally significant secondary structure elements
(SSEs) of the pol domain of Pol γA, including helixes αO, αL and αQ, as well as the loop
corresponding to the 7β-loop-8β motif in T7 Pol, are tilted away from the active site, as compared
to the analogous SSEs from T7 Pol and Klenow. This suggests conformational rearrangements in
the Pol γ catalytic core upon DNA binding. To model the ptDNA in the putative DNA-binding
channel of Pol γA, and to identify residues that interact with the DNA duplex, we performed a
comparative analysis of the T7 Pol (24) and Klentaq (26) DNA complexes in both the open and
closed states. We sought to identify SSEs in the pol domain that fulfill the requirements of
59

interaction with DNA and preservation of conserved positions between the enzymes. As a result,
the Q-helix and the 7β-loop-8β motif were identified, and a third alignment of Pol γA and T7 Pol
was based on superposition of their αQ helices (residues M1093-E1122 in Pol γ and P606-E635
in T7 Pol), and the hairpins formed by the 7β-loop-8β (residues T846-V855 in Pol γ and P422T431 in T7 Pol). The RMSD for the superposed hairpins was 1.397 Å.
Clustering of Alpers disease mutations within the catalytic core of Pol γ
Recessive mutations are distributed along the length of the POLG1 gene sequence
(http://tools.niehs.nih.gov/polg/), but they cluster within the tertiary structure of Pol γA to distinct
regions within the catalytic core (Figure 22 on page 62). Five functional modules, termed clusters
1–5, were assigned within Pol γA as follows: Cluster 1 (in green) locates within the pol domain
and comprises largely residues affecting DNA polymerase activity; Cluster 2 (in yellow)
represents residues lining the upstream DNA-binding channel; Cluster 3 (in red) is associated with
a novel structural motif in the fingers subdomain, which we propose confers partitioning of the
DNA substrate between the pol and exo sites; Cluster 4 (in blue) lies on the periphery of the exo
domain and mediates interactions with the distal Pol γB, stabilizing the Pol γ: DNA complex; and
Cluster 5 (in cyan) locates within the spacer domain and represents a region that we propose is
involved in replisome interactions.

60

Figure 22. Alpers disease mutations cluster within functional modules in the catalytic subunit of Pol γ.

61

Figure 22 (cont’d).

62

Figure 22 (cont’d) Upper panel: schematic diagram of the POLG1 gene showing the distribution
of recessive Alpers disease mutations (Human DNA Polymerase gamma Database,
http://tools.niehs.nih.gov/ polg/).AID and IP in the spacer domain refer to the accessory (subunit)
interacting and intrinsic processivity subdomains, respectively, that are discussed in the text;
NTD refers to the N-terminal domain. Lower panel: tertiary structural representation of the
apoenzyme form of Pol γ [PDB code 3IKM, (7)] with modeled DNA, identifying the positions of
five functional modules (shown in mesh) that are defined by clusters of amino acid residues
(shown as spheres) affected by Alpers disease mutations as follows: Cluster 1, green; Cluster 2,
yellow; Cluster 3, red; Cluster 4, blue; Cluster 5, cyan. The domains of Pol γA are shown as
surface representations, and in part as secondary structural elements (SSEs) that are colored as
depicted in A. The proximal and distal accessory subunits are shown as surface representations
in light and dark gray, respectively. Primer–template DNA was docked as described in the text
and is displayed as orange ribbons.

63

RESULTS
ALPERS CLUSTER 1: Residues affecting 5′-3′ DNA polymerase activity
Catalysis of template-directed nucleotide polymerization in DNA polymerases proceeds
via a phosphoryl transfer reaction assisted by two Mg2+ ions (27, 28). Incorporation of the correct
dNMP is mediated by a nucleotide-induced conformational shift of the fingers subdomain between
an open and closed state; the closed state results in phosphoryl transfer when correct Watson–
Crick base pairs between the incoming nucleotide and a templating base can fit stably in the active
site (29, 30). Similarities in the overall mechanism of DNA polymerases suggest a conserved
structure of their active sites and provide a basis for comparative structural analysis and
extrapolation of site-directed mutagenesis data from one DNA polymerase to its close homologs.
The amino acid residues that we attribute to Alpers Cluster 1 contribute to nucleotide
polymerization per se, to shaping the overall architecture of the pol active site, or to positioning
the primer template within the active site (Figure 23A on page 72).
Catalytic residues
The largest group of recessive Alpers mutations map to the pol domain in POLG1 (Figure
22 on page 62), and a number of the affected amino acids are likely involved in catalysis by Pol γ.
Residues E1136 and K1191 surround the highly conserved residues D890 and D1135 (Figure 23B
on page 72), which locate at the base of the pol active site and have been shown to coordinate the
two magnesium ions required for catalysis (7). In vitro mutagenesis of residue E883 in Klenow
(E1136 in Pol γ) resulted in decreased pol rate, kcat and affinity for dNTPs and DNA (31, 32),
which fits with the recessive nature of this mutation and its effect on mtDNA copy number in vivo
(16, 23). Comparative structural analysis of Pol γ and its close homologs revealed that E895 likely
confers discrimination against ribonucleotides, and it is involved in aligning the incoming dNTP
64

within the catalytic site, together with the catalytic residues Y955 and Y951 [Figure 23C on page
72 (33)]. Interestingly, E895 also occupies part of the dNTP-binding pocket, and whereas E895G
is a recessive mutation, Y955C is dominant. In vitro mutagenesis of Pol γ showed that the Y955C
variant exhibited dramatically reduced catalytic activity, but had the same affinity for DNA as the
wild-type enzyme (33). In a disease situation, these effects cannot be compensated for by the wildtype allele and as a result, Y955C causes a dominantly inherited adult-onset mitochondrial
myopathy (autosomal dominant progressive external ophthalmoplegia) associated with mtDNA
deletions (34). E1136 is part of one of three conserved sequence motifs in the palm subdomain of
family A DNA polymerases (the PolC motif, Figure 23B on page 72). Its counterparts E655 in the
T7 Pol (25) (Figure 23B on page 72) and E786 in the Klentaq (26) structures are not in contact
with primer or incoming dNTP, but mutagenesis of Klenow showed that this residue is critical for
catalysis (31). The ternary complex of T7 Pol reveals an electrostatic interaction between E655
and H704, and suggests that the role of E655 is to position H704 in close proximity to the
phosphate moiety of the 3′ nt in the primer strand, and it was shown to be essential for catalysis
(25). K1191 of Pol γ occupies the equivalent position of H704 in T7 Pol, suggesting that it would
interact with E1136 in a similar manner upon binding of the primer template to facilitate catalysis.
In contrast to the rigid pocket formed by the palm active site residues, the opposite half of
the active site is highly flexible to allow a fast transition between open and closed conformations
of the fingers subdomain. This conformational asymmetry provides the structural basis for the high
affinity of DNA polymerases for the primer–template junction: the template must be bound firmly
and consistently each catalytic cycle to assure the fingers sufficient freedom to examine each
incoming nucleotide. Several Alpers mutations have been reported in this region of the fingers,
scattered mainly along the O-helix (Figure 23C on page 72). Biochemical studies on the A957S

65

mutant showed mild defects in kcat and dNTP binding, but a modest increase in DNA-binding
affinity (33). A957 locates in the loop between helices O and O1, within one of the three hinge
regions, the ‘GAG’ hinge (residues G956/A957/G958), which in related DNA polymerases is
known to confer the flexibility necessary for closing and opening of helixes O and O1 (27,35).
Glycine to alanine substitution in the GAG hinge in T7 Pol resulted in complete loss of catalytic
activity, which can be attributed to complete or diminished ability to shift between the open and
closed states (36). Mutations affecting the hinge region likely introduce steric clashes that would
limit the extent of conformational shifts. Interestingly, different amino acid replacements for A957
in Pol γ do not have similar consequences: A957S causes dominantly inherited adult-onset
myopathy with multiple mtDNA deletions, whereas an A957P substitution has been reported only
as a recessive, compound heterozygous mutation in patients with Alpers syndrome and mtDNA
depletion. A957S showed an increased affinity for DNA (33), a consequence that is consistent
with a slower open–close rate, and could potentially cause pol stalling or dissociation. In contrast,
the severe defect caused by A957P suggests that the presence of proline in the hinge abolishes the
flexibility between helixes O and O1, a defect that we predict will result in decreased affinity for
both dNTP and DNA and may prohibit the formation of a productive ternary complex. The loop
between helixes O and O1 has been shown to form the pre-insertion site in Geobacillus
stearothermophilus Klenow to accommodate the ssDNA template and prevent its premature entry
into the active site (35). However, the available library of DNA polymerase structures reveals
variability in the pre-insertion site, making it difficult to assign a critical role. Residue K947, which
is associated with Alpers syndrome, is located on the O-helix adjacent to R943. The corresponding
residues in T7 Pol (K522 and R518, respectively) contact the triphosphate moiety of the incoming
nucleotide in the closed ternary complex (PDB code 1T8E). These electrostatic interactions
66

provide most of the driving force for O-helix movement, and mutations in these positions will
weaken interactions.
Architectural residues
Optimal catalysis requires not only the residues interacting directly with dNTP and primer
template, but also those that maintain the architecture of the pol active site. We suggest that the
pathogenic role of the Alpers mutations Q879H, T885S, L886P, G888S, E1143 and D1184N can
be attributed to modifications in the local β-sheet architecture surrounding the catalytic residues in
the palm (Figure 23A, upper panel on page 72). Defects are predicted to reduce pol rate, and it
appears that the closer the mutated architectural residue is to the catalytic site, the more pronounced
its effect. In support of this, in vitro mutagenesis of Q879 and T885, located a distance of ~15 Å
from the catalytic residues, caused only a 2-fold decrease in pol rate (20). In contrast, mutagenesis
of E883 located at the beginning of the strand β13 within the catalytic site in Klenow (E1136 in
Pol γA) reduced kcat 26-fold (31).
The Alpers mutation D930N (Figure 23C on page 72) was studied in vivo in yeast, and led
to complete loss of mtDNA (37). The corresponding residue in T7 Pol (D504) interacts with
triphosphate-binding residue R518 (R943 in Pol γ) in the closed conformation (PDB code 1T8E),
and therefore contributes indirectly to correct dNTP binding. Residues T914 and L966 located in
the fingers subdomain (Figure 23C on page 72) are conserved only among Pol γs, and are linked
to Alpers syndrome. Our structural analysis predicts that mutations in these residues will likely
compromise the stability and conformation, and/ or the transition rate between the open and closed
states of the fingers. In sum, we propose that the primary effect of mutations surrounding the Ohelix residues is altered affinity for dNTPs, which decreases pol rate.

67

Residues conferring a high affinity of the pol site for the primer–template junction.
Docking of DNA onto the Pol γ structure shows that the DNA-binding channel of the
enzyme encompasses ~20 bp of DNA duplex, whereas those in T7 Pol and Klentaq are relatively
short and interact with ~8 bp of DNA duplex (24–26). In Pol γ, similar to other family A DNA
polymerases, the DNA-binding channel starts with residues located in the β7-loop-β8 motif in the
palm, which form extensive contacts with the first four base pairs in the ptDNA. In Pol γA, this
fragment encompasses residues 842–856 and appears as a loop in the crystal structure with the
current resolution of 3.24 Å (7). Because the β7-loop-β8 motif has two arginines in its tip, R852
and R853, the former of which is conserved only among Pol γs, we designated it as the ‘RR loop’
(Figure 23D on page 72). Several recessive mutations affect the RR loop: G848S, T849H, T851A,
R852C and R853Q, as well as the surrounding residues H1134R and H1110Y. In vitro mutagenesis
of G848, T851, R852 and R853 in human Pol γ (20), and R668 in Klenow ((31, 32), corresponding
to R853 in Pol γ), caused a dramatic decrease in catalytic activity and DNA-binding affinity.
Polesky et al. (32) highlighted the complex effect of these mutations, noting that a specific amino
acid can bind a critical part of the substrate, contributing to the overall affinity of the enzyme for
ptDNA, and to catalysis by positioning the substrate in the active site. Comparative structural
analysis of Pol γ homologs solved in ternary complexes with bound DNA prompts us to propose
that the primary function of the RR loop is in binding and alignment of ptDNA within the DNAbinding channel. H1134 locates adjacent to the RR loop in the tertiary structure (Figure 23D on
page 72). The corresponding histidines in Klenow (31, 32), T7 Pol (25) and Klentaq (26) were
shown to be essential for DNA binding, and for coordination of the first nucleotide in the primer
strand. Analysis of Pol γ suggests that Q1102 from the Q-helix (Q615 in T7 Pol) and R853 (R429
in T7 Pol) from the RR loop coordinate the first base pair in the ptDNA, and therefore may be
crucial in sensing a mispair. T851 (T427 in T7 Pol) likely participates indirectly by positioning
68

Q1102. T849 is predicted to coordinate the DNA duplex by interaction with the phosphate moiety
of the third nucleotide from the 3′-end of the primer. In the Pol γ apoenzyme structure, the side
chains of the residues from the RR loop are oriented differently than the orthologous residues in
the T7 Pol ternary complex, which suggests that DNA and dNTP binding affect the positions of
the residues that coordinate them. For example, R853 in the Pol γ apoenzyme interacts with the
metal-binding residue D1135 as was noted by Stumpf and Copeland (17) and with E895, but in
complex with DNA, R853 would shift to interact with the base of the 3′ nt of the primer strand, as
in T7 Pol (24–26). A distinguishing feature of the RR loop, in comparison with the orthologous
β7-loop-β8 motif in other family A DNA polymerase group members, is the electrostatic
interaction between R852 and D1107, which may confer additional stability to the RR loop in Pol
γ (Figure 23D on page 72). Disease mutations in this domain affect primarily DNA binding and
positioning in the active site channel, manifesting defects in both DNA-binding affinity and in
catalytic efficiency.

69

Figure 23. Alpers Cluster 1 mutations affect the 5 DNA polymerase activity of Pol γ.

70

Figure 23 (cont’d)

71

Figure 23 (cont’d)

Amino acid residues affected by Alpers Cluster 1 mutations in POLG1 are shown as green
spheres. Other Pol γ residues that are discussed in the text are shown in brown, and T7 residues
are shown in red. Pol domain SSEs are shown in pink according to the schematic shown in
Figure 22 on page 62; ptDNA is indicated by orange (template) and brown (primer) strands. The
incoming dNTP is shown in blue. Mg2+ ions are shown as small gray spheres. A, upper panel,
overview of the positions of Alpers Cluster 1 mutations with dashed black lines indicating the
regions described in the text and depicted in B–D; A, lower panel, overview of the positions of
Alpers Cluster 1 mutations relative to ptDNA and incoming dNTP; B–D, positioning of Pol γ
residues in the apoenzyme form [PDB code 3IKM (7)] and T7 Pol residues in its closed ternary
complex [PDB code 1T8E (25)] relative to ptDNA and incoming dNTP, with the dashed arrow
showing the expected movements of the Pol γ residues upon formation of a closed complex: B,
Alpers mutations surrounding the Mg2+ -binding residues; C, Alpers mutations affecting O-helix
movement and dNTP binding, with the T7 Pol O-helix in its closed conformation superimposed
in red; D, Alpers mutations affecting the RR loop and the surrounding residues.
CLUSTER 2: Recessive mutations affecting the upstream DNA-binding channel
Processivity of a DNA polymerase depends both on DNA-binding affinity and on the rate
of DNA polymerization. Structural analysis shows that the mutations affecting interaction of Pol
γ with the upstream DNA duplex are located within the spacer domain (Figure 22 on page 62).
This would argue that the AID [accessory (subunit)-interacting domain] and IP (intrinsic

72

processivity) subdomains adopt a different conformation in Pol γ: DNA complexes. To evaluate
the potential impact of recessive spacer region mutations on DNA binding, we considered results
of in vitro mutagenesis of spacer domain in Drosophila Pol γ (38) and in Saccharomyces cerevisiae
Mip1 (22) (Figure 24 on page 74). Analysis of the Pol γ structure with modeled ptDNA reveals
that residues K755 from the IP subdomain and Q497 in the K-tract of the AID subdomain interact
with upstream DNA. In the human Pol γ: DNA model, the L752P and A767D substitutions would
affect the structure in the tip of IP subdomain, which contacts the minor groove of the DNA duplex
in a similar way as reported for T7 Pol [residues T354-V363 (25)]. In the Pol γ: DNA complex,
L752 and A767 neighbor the K768/D769/F770 triplet from fly Pol γ; triple alanine substitution of
KDF in fly Pol γ resulted in a 1.4-fold decrease in DNA-binding affinity (38). We predict that the
Alpers A767D mutation likely causes similar consequences for DNA binding. Mutations of P587
and P589 in the spacer domain are also linked to Alpers disease. These residues constrain the β–
hairpin loop between the IP and AID subdomains and are positioned close to the Y452/E453/E454
triplet that lies within the thumb [corresponding to the YED triplet in Drosophila Pol γ (38)]. We
postulate that the P587L and P589L mutations may suffer the same consequences as for the triple
alanine substitution of YED in the fly enzyme, which caused reduced processivity, most likely as
a result of misalignment of the ptDNA, with respect to the pol catalytic site (38). Furthermore, a
triple alanine mutant of P556A/K557A/L558A in the fly was shown to be completely deficient in
DNA binding and pol activity (38); this defect is likely due to disruption of the hydrophobic
structure of the spacer domain, which shapes the DNA-binding channel wall. Notably, the residue
affected in the common A467T human disease allele is also part of this PKL hydrophobic center,
and its biochemical properties most likely derive from similar structural alterations that interrupt
hydrophobicity with the introduction of a hydroxyl group. Whereas we and others have found that

73

the A467T catalytic core alone exhibits highly reduced DNA-binding affinity and pol activity (19,
21), we showed that these defects are mitigated partially by its association with the accessory
subunit (21). This likely results from partial stabilization of the overall conformation of mutant
catalytic core within the reconstituted Pol γ holoenzyme.

Figure 24. Alpers Cluster 2 mutations affect the upstream DNA-binding channel of Pol γ.
Amino acid residues affected by Alpers Cluster mutations in POLG1 are shown as yellow
spheres. Other Pol residues that are discussed in the text are shown in brown. Spacer domain
SSEs are shown in magenta and pol domain SSEs are in pink according to the schematic shown
in Figure 22 on page 62; ptDNA is indicated by orange (template) and brown (primer) strands.
The spacer domain is also shown as a transparent surface representation in pale gray and the exo
domain is shown in purple.
CLUSTER 3: Mutations associated with a novel, Pol γ-specific functional module,
conferring partitioning of DNA substrate between the pol and exo active sites
Comparative analysis of the pol domains from human Pol γ, T7 Pol and T7 RNA Pol
reveals a remarkable structural similarity in their overall folds. In fact, the only significant
difference between the analogous pol domains lies in the region that connects the P-helix of the
74

fingers subdomain and Q-helix of the palm (see Figure 30 in appendix B on page 92), which
contains a novel module (Figure 25 on page 78) whose amino acid sequence is highly conserved
in Pol γs from yeast to man. This, and the finding that several recessive Alpers mutations cluster
in this region (Figure 22 on page 62), argue that it serves an important role. Indeed, in our Pol γ:
DNA model, this module extends into the DNA-binding channel in a position very near the pol
active site (<10 Å from the Mg2+-binding catalytic residues). In bacteriophage N4 RNA
polymerase (PDB code 2PO4) (39) and in T7 RNA polymerase [PDB code 1ARO (40); Figure 30
on page 92], the fragment connecting the fingers with the helix corresponding to the Q-helix in
Pol γ participates in specific recognition of the transcriptional promoter; it is aptly termed a
specificity loop (39–41). In Pol γ, we have adopted the term ‘partitioning loop’ based upon the
functional role we propose it serves. We discuss our rationale below, and use it to establish
structure–function justifications of disease manifestation for the Alpers mutations we assign to
Cluster 3.
We propose that the partitioning loop modulates the partitioning of the primer strand
between the pol and exo active sites by forming stable contacts with correctly base-paired ptDNA,
and destabilizing ptDNA that contains mispairs or lesions. Comprising residues 1050–1095, the
atomic structure reveals that the first 10 residues of the partitioning loop adopt an α-helical fold
that extends from the fingers subdomain directly into the DNA-binding channel. The beginning of
this helix contains a W1049XGG1052 motif, which is strictly conserved in Pol γ, and a similar
motif is also present in T7 Pol (W579XAG582). Our Pol γ: DNA model shows that the WXXG
motif maps to the same region of the fingers domain in both Pols. In T7 Pol, it is responsible for
binding downstream template ssDNA via base-stacking interactions between W579 and the
nucleotide base [PDB code 1T8E (25)], and likely performs an equivalent function in Pol γ. The
75

novel functionality that the partitioning loop contributes to Pol γ is achieved by its unusual loop–
hairpin component spanning residues 1060–1074 that appears to grip the modeled ptDNA along
the major groove as shown in Figure 25A on page 78. We propose that it uses a steric exclusion
mechanism that confers exquisite specificity by associating closely with the major groove of
primer template, such that correctly base-paired substrates are bound stably. In contrast, DNA
lesions and mispairs that adopt an altered helical structure clash sterically with the partitioning
loop, resulting in exclusion of the primer template from the pol active site. Excluding primer
template may facilitate fraying of the duplex, generating ssDNA primer ends that can be bound by
and hydrolyzed in the exo active site (Figure 25B on page 78).
Mutations in the residues composing the partitioning loop, such as the Alpers mutations
R1047W and P1073L, as well as the progressive external ophthalmoplegia mutations G1051R and
G1076V, have been studied in vivo. Yeast strains homozygous for G1051R Pol γ exhibited a point
mutational frequency >10-fold higher than the wild-type strain, and heterozygous strains showed
frequencies >2-fold higher relative to homozygous wild-type strains (42). Increased point mutation
frequencies were also reported for yeast strains heterozygous and homozygous for the Alpers
mutation P1073L and the G1076V mutation associated with progressive external ophthalmoplegia
(22, 43). Both the P1073L and G1076V mutations cause substitutions in strictly conserved residues
located in the loop region that is expected to contact DNA, and the G1051R mutation affects a
strictly conserved glycine in the WXXG motif. Interestingly, yeast strains heterozygous and
homozygous for L304R, S305R, Q308H, R309L and R309H showed similar increases in point
mutation frequency as well as mtDNA depletion (43). These exo domain residues are located in a
loop–helix motif adjacent to the partitioning loop, and have been described previously as the
orienter module (22) (Figure 25 on page 78). When S305R and P1073L were present as compound

76

heterozygous, the point mutation frequency increased drastically to >70-fold that of the wild-type
strain (44). Non-complementation of these two mutations suggests that both are involved in the
same function, and we propose that the role of the orienter module is to position correctly the
partitioning loop. L304R, R309H, R309L and W312R were analyzed biochemically by producing
and purifying homologous yeast Pol γ variants (22). All variants exhibited reduced DNA-binding
affinity and reduced pol activity and in addition, the L304R variant showed a significant increase
in exo activity (22). An identical biochemical phenotype was reported previously in the fly Pol γ
SYW triple alanine variant, which correspond to thumb residues S799/F800/W801 in human Pol
γ (38). As illustrated in Figure 25 on page 78, the SYW residues are located on the face of the
thumb subdomain directly across the DNA-binding channel from the orienter module, and we
predict that this region of the thumb plays an equivalent role to the orienter module. Therefore, we
define Cluster 3 as mutations affecting residues of the partitioning loop, as well as nearby residues
that govern the position and conformation of the partitioning loop. Mutations in Cluster 3 will alter
the balance between pol and exo activity and diminish the fidelity of the polymerase. We note the
possibility that enhanced fidelity may also be experimentally feasible, introducing the potential to
engineer fine-tuned anti-mutator replicase.

77

Figure 25. Alpers Cluster 3 mutations are associated with a novel Pol γ-specific functional
module proposed to be involved in primer strand partitioning between the pol and exo
sites. Amino acid residues affected by Alpers Cluster 3 mutations in POLG1 are shown as red
spheres/ mesh adjacent to a novel alpha helix with an associated loop–hairpin (the partitioning
loop), also shown in red. The brown spheres and mesh represent the SYW (fly)/ SFW (human)
and surrounding residues, respectively, that are described in the text. The pol domain is shown as
a surface representation in pink and the exo domain is shown in purple, according to the
schematic shown in Figure 22 on page 62; ptDNA is indicated by orange (template) and brown
78

Figure 25 (cont’d) (primer) strands. (A) The predicted position of the partitioning loop relative to
ptDNA in the pol mode, and (B) represents the exo mode. To dock the frayed ptDNA in the exo
active site, the exo domain residues 324–518 of Klenow (PDB code 1KLN) were aligned with
the exo domain residues 170–440 of Pol γ (PDB code 3IKM) (see appendix B for Figure 31 on
page 94).
In sum, our justification for the proposed role of the partitioning loop is based on its
sequence conservation among Pol γs from yeast to man, its positioning relative to the specificity
loop found in T7 RNA Pol, the clustering of several recessive Alpers mutations in this region, its
absence in any other Pol family and the documented high fidelity of Pol γ. We argue that an
alternate hypothesis that this structural element would simply rotate away upon DNA binding (7)
seems unlikely based on the above considerations and, in particular, on the evolutionary
conservation of a novel structural module that is positioned strategically in the DNA-binding
channel very near the pol active site. Clearly, validation of either hypothesis warrants future
experimentation.
CLUSTER 4: Mutations affecting Pol γA interactions with the distal Pol γB upon DNA
binding by Pol γ holoenzyme
Pol γA contains independent 5′–3′ DNA polymerase and 3′–5′ exonuclease active sites
whose functions are coordinated in proofreading DNA synthesis. The human Pol γ crystal structure
showed that its homodimeric Pol γB interacts asymmetrically with Pol γA, such that one Pol γB
protomer forms the dominant subunit interface and is designated as the proximal accessory
subunit, while the other protomer makes very limited contact with Pol γA and is designated as the
distal accessory subunit (7). In fact in the apoezyme structure, the Pol γA interaction with the distal
Pol γB is mediated through one ion bond between Pol γA R232 and Pol γB E394 [(7) and Figure
26 on page 81]. A subsequent biochemical study on the interaction between the catalytic and
accessory subunits showed that the proximal Pol γB increases DNA-binding affinity of the
holoenzyme, while the distal protomer enhances the polymerization rate of the holoenzyme (45).
79

Again as suggested in the section on Cluster 2, we posit that the holoenzyme likely undergoes
substantial conformational rearrangements in the IP and AID subdomains of the spacer upon
formation of an active complex with DNA.
Mapping of Alpers mutations within the crystal structure revealed that the majority of the
mutations in the exo domain cluster on the protein surface within 13–17 Å from Pol γA residue
R232 (Cluster 4 shown in blue in Figure 22 on page 62 and Figure 26 on page 81). The functional
role of this region can be deduced from biochemical studies on mutations of the neighboring R232
(46) and L244 residues (22, 43). Mutations of R232 were shown to decrease pol activity, DNAbinding affinity and processivity of the holoenzyme yet at the same time, to enhance its
exonuclease activity, which was also rendered less selective for mismatches (46). These data argue
that in the Pol γ: DNA complex, and/or in its ternary complex with dNTP, the distal accessory
subunit associates more tightly with the catalytic subunit, thereby enhancing binding of the
upstream DNA in the pol mode, and limiting the rate of translocation of the frayed 3′-end of the
primer to the exo site (46). Accordingly, an orthologous variant of the L244P human mutation in
yeast caused increased mutation frequency (22, 43). Previous site-directed mutagenesis of human
Pol γB also suggested extensive interaction between the catalytic and distal accessory subunits
within the ternary complex (5). Analysis of the Pol γ apoenzyme structure shows that E445 and
T447 of Pol γB are located at the tip of the its C-terminal region, and that only in the distal subunit
are these residues oriented toward Cluster 4 on the edge of exo domain. A double mutation of these
residues (E445A/ T447A) led to a dramatic decrease of the stimulatory effect of Pol γB on the
wild-type Pol γA, and a decrease in DNA binding (5). We propose that residues in Cluster 4 of Pol
γA contribute to tight interaction between exo domain of catalytic subunit and the distal accessory
subunit, resulting in more extensive enclosure of the primer template in the pol mode. We also

80

predict the participation of the AID of the spacer domain in these conformational changes, because
it is associated intimately with the accessory subunit. In support of this, biochemical studies
performed on the yeast Pol γ core variant of Alpers mutation R574, which is located in the AID
subdomain, showed not only reduced DNA-binding affinity, but also reduced exo activity and a
severe processivity defect (22, 43). To some extent, these interactions would provide structural
constraints on the upstream DNA in the direction toward the partitioning loop. Therefore, the
partitioning loop, orienter module and thumb in Pol γA, and the interface between its exo domain
and the distal accessory subunit might function in a concerted manner to affect switching between
the pol and exo modes in Pol γ function.

Figure 26. Alpers Cluster 4 mutations affect Pol γA interactions with the distal Pol γB
upon DNA binding by Pol γ holoenzyme. Amino acid residues affected by Alpers Cluster 4
mutations in POLG1 are shown as blue spheres/ mesh and are located largely within the exo
81

Figure 26 (cont’d) domain (shown in purple). The helix shown in magenta is the AID (accessory
interacting domain) of the spacer region that interacts with the proximal accessory subunit. Other
domains are colored according to the schematic shown in Figure 22 on page 62; duplex DNA is
indicated by the orange strand. Orange spheres indicate the positions of the accessory subunit
residues described in the text. The proximal and distal accessory subunits are shown as surface
representations in light and dark gray, respectively.
CLUSTER 5: Mutations affecting a region of the IP subdomain that is likely involved in
replisome contacts.
Many reported Alpers mutations (L623W, R627W/Q, P648R, G737R, G746S, W748S and
F749S) map to the distal surface of the IP subdomain (Figure 27 on page 83). We considered these
as distinct from the Cluster 2 residues because they are located much further from the DNAbinding channel. In addition, in vitro studies of W748S and R627W/Q variants showed no defects
in pol activity, processivity or DNA-binding affinity (47, 48). In contrast, biochemical defects
were demonstrated earlier in single alanine variants of fly Pol γ within this region of the spacer: a
W576A variant was nearly inactive, F578A retained half of wild-type activity, G575A displayed
wild-type activity and all three variants showed substantially reduced stimulation by mtSSB (38).
The fly G575/W576/F578 residues correspond to residues G619/W620/Y622 in human Pol γ. In
the human Pol γ apoenzyme structure, W620 is buried in the hydrophobic core of the IP
subdomain, whereas Y622 and G619 are located closer to the surface (7). The combined
biochemical data suggest that residues closer to the distal surface of the IP domain are not critical
for catalytic function. Cluster 5 mutations map to this surface, and therefore a catalytic defect per
se is not expected to be the cause of Alpers manifestation. Rather, we propose that this region is
involved in protein–protein contacts, a likely partner being mtSSB as suggested by the GWF fly
Pol γ variants (38). Another candidate may be the mitochondrial replicative DNA helicase mtDNA
helicase, but such an interaction has not been investigated to date.

82

Figure 27. Alpers Cluster 5 mutations are proposed to affect replisome interactions. Amino
acid residues affected by Alpers Cluster 5 mutations in POLG1 are shown as cyan spheres.
Spacer domain SSEs within the IP (intrinsic processivity subdomain) are shown in magenta;
ptDNA is indicated by orange (template) and brown (primer) strands. The spacer domain is also
shown as a transparent surface representation in pale gray and the exo domain is shown in
purple. Brown spheres indicate the positions of the IP residues described in the text.

83

Table 5. Structural and functional features of the five proposed Alpers Clusters.

Predicted Primary
Biochemical Defect

Predicted Phenotype
(primary, secondary)

Causes
Alpers when
found in
trans with
cluster

1

Pol active site
and environs

Decreased pol rate

Reduced pol rate
Reduced DNA binding
Reduced processivity

2, 3, 5

2

Upstream
DNA-binding
channel

Decreased DNA
binding affinity

Reduced DNA-binding
Reduced pol rate
Reduced processivity

1, 3, 4, or 5

3

Partitioning
loop

Partitioning of primer
strand between pol
and exo active sites

Altered exo/ pol ratio
Altered error rate
Altered exo rate
Reduced DNA-binding
Reduced processivity

1, 2, 5

4

Exo-IP
interface with
POLGB

Stabilization of
ternary complex

5

Periphery of IP
domain

Protein–protein
interactions?

Cluster

Structural
location

84

Reduced pol rate
enhancement by Pol γB
Increased error rate
Reduced processivity
Reduced SSB stimulation
Deficient protein-protein
interactions?

2 or 5

1, 2, 3, 4,

PROSPECTS
Toward a diagnostic tool to assess new human mitochondrial disease mutations
We show that although the reported Alpers mutations scatter across the entire POLG1 gene,
they cluster to distinct functional regions in the Pol γ tertiary structure. Table 5 on page 84
summarizes the structure–function predictions for each cluster and also displays the cluster
combinations that have been reported in compound heterozygous patients with Alpers syndrome.
These combinations do not occur randomly; rather, multiple occurrences of specific cluster
combinations is evident whereas others are absent, suggesting that the latter combinations are not
tolerated. Cluster 1 mutations locate in the pol active site region and will invariably cause reduced
pol activity, potentially combined with reduced DNA binding. Cluster 2 mutations locate in the
upstream DNA-binding channel, resulting in reduced DNA-binding affinity, but are too distant
from the catalytic site to affect directly pol activity to the extent of Cluster 1 mutations. Therefore,
Cluster 2 mutations will be recessive, whereas Cluster 1 mutations may be either dominant or
recessive, depending on the severity of the biochemical defect. Cluster 3 mutations locate to the
partitioning loop or its environs, such as the orienter module, and we predict that they will alter
the balance of polymerase and exonuclease activities of mutant Pol γ. Cluster 4 mutations locate
on the surface of the exo domain along the interface of the distal accessory subunit. These
mutations are predicted to reduce the stimulation affected by the distal accessory subunit, which
has been documented to enhance pol rate and to reduce exo activity (45). Both Clusters 3 and 4
mutations will likely cause increased mutagenesis in vivo, a phenotype that has been observed in
yeast models to be associated with amino acid alterations within these clusters (43). Cluster 5
mutations locate on the distal surface of the intrinsic processivity (IP) subdomain, removed from
the pol active site, DNA-binding channel and accessory subunit interface. Their distant location
and lack of biochemical defects prompt us to speculate that this region is involved in replisome
85

contacts. This hypothesis, in particular, highlights the need for future experimentation on both the
physical and functional interactions among the key proteins at the mtDNA replication fork.
POLG1 shows significant polymorphic variation in the human population, and a constant
challenge in DNA diagnosis is to distinguish pathogenic mutations from neutral variants. We
suggest that the remarkable clustering of Alpers mutations into specific functional regions (Figure
22 on page 62) enables the use of our Pol γ: DNA model as a diagnosis-supporting tool for
evaluating pathogenic potential of new sequence variants. For example, Q1102 is located within
the Cluster 1 region, and mutations affecting this amino acid can be predicted to cause catalytic
defects. This prediction is consistent with biochemical studies on the Q1102 counterpart in
bacteriophage T7 Pol (Q615), which was shown to be essential for the fidelity of nucleotide
incorporation and for ptDNA binding (25). Functional predictions can be made in the absence of
biochemical data as well; for example, the tam9 mutant allele, carrying an E595V mutation in the
catalytic subunit of Pol γ, causes mtDNA depletion in the fly (49, 50). E595V corresponds to
E641V in human Pol γ, which maps to Cluster 5 (Figure 27 on page 83). Our model predicts that
the tam9-associated mtDNA depletion is caused by deficient replisome interactions. Consequently,
we would predict that both Q1102 and E641 are potential candidates for causing recessive POLG
syndrome with mtDNA depletion.
Predicting the consequences of compound heterozygosity via Alpers cluster analysis
Alpers syndrome is typically caused by compound heterozygosity of two mutations
occurring along the entire length of the POLG1 gene. In contrast, homozygosity for a single
mutation often is associated with a somewhat milder disorder (2, 51). Utilizing the Human DNA
Polymerase γ Database, we compiled data to show all combinations of mutations that have been
reported to trigger Alpers disease (Figure 28 on page 88). Alpers patients typically do not show a
86

combination of two mutations from the same cluster. In addition, Cluster 4 mutations only manifest
as Alpers disease when combined with Clusters 2 or 5 mutations. These trends provide merit for
our structure-guided assignment of the mutations into the five proposed clusters, and suggest
unique functional roles inherent to each. With the rapid development of massive parallel
sequencing methodologies, the number of new Pol γA variants is likely to increase significantly,
emphasizing the importance of bioinformatic tools to evaluate the pathogenic role of identified
variants. Furthermore, the identification of synergistic functional relationships between the
characterized clusters provides novel insight into the mechanism of Alpers disease manifestation,
and will serve as a framework for future structure–function studies of the mitochondrial DNA
replicase.

87

Figure 28. Combinations of mutations found in Alpers patients. Individual Alpers mutations
are grouped by cluster as shown in Figure 22 on page 62, and black blocks represent Alpers
88

Figure 28 (cont’d) manifesting combinations. The inset in the lower right presents a simplified
version of the table by reducing the axes to the five clusters only, where gray blocks represent
cluster combinations that have not been found in Alpers patients. The tabulated data suggest that
two mutations from the same cluster do not typically manifest as early-onset Alpers disease.
Furthermore, the data indicate that Cluster 4 (blue) mutations manifest as Alpers disease only
when in combination with Cluster 2 (yellow) or Cluster 5 (cyan) mutations. These trends support
the existence of unique functional relationships in Pol γA that are inherent to each cluster.

89

APPENDIX

90

APPENDIX

Figure 29. Alignments used for docking DNA onto the Hs Pol γ holoenzyme structure
(3IKM). Panels display the palm subdomain of the T7 Pol (1T8E) closed ternary complex (red)
superimposed on the palm subdomain of Pol γ (pink). Top panel, alignment of the palm of T7
Pol (residues 409-487, 611-704) to the palm of Pol γ (815-910, 1095-1239); middle panel,
alignment of T7 Pol residues 646-663 (green) to Pol γ residues 1127-1144; lower panel,
alignment of T7 Pol residues 606-635 and 422-431 (green) to Pol γ residues 1093-1122 (Q-helix)
and 846-855 (RR-loop). Alignments were performed in Pymol.
91

Figure 30. Comparative alignments of Hs Pol γ, T7 Pol and T7 RNA Pol illustrating the
structural variations in the region between the P-helix and the Q-helix in the pol domain.
Panels on the left display the complete pol domain in complex with DNA, and those at right
92

Figure 30 (cont’d) show a close up view of the of the region between the Q-helix and P-helix.
Top panel, the Pol γ (3IKM) pol domain is shown as pink cartoon and its partitioning loop in red,
with its disordered region indicated as a dashed red line. Transparent surfaces of the exo (purple)
and pol (pink) domains are also shown at left. DNA docked by alignment with T7 Pol (see top
panel of Fig. 29 on page 91) is shown in orange; middle panel, the Pol domain of the T7 RNA
Pol elongation complex (1H38) is shown as light blue cartoon and its specificity loop in red;
lower panel, the T7 Pol (1T8E) pol domain is shown as blue cartoon, and transparent surfaces of
its exo (purple) and pol (blue) domains are shown to highlight a similar architecture as in Pol γ.
WXGG and WXAG are the amino acid sequence motifs in Pol γ and T7 Pol, respectively, which
are discussed in the text.

93

Figure 31. Structural alignment of the exo domain of Klenow editing complex (PDB code
1KLN, residues 324-518, displayed as blue cartoon) with the exo domain of human Pol γ
(PDB code 3IKM, residues 170-440, displayed as purple cartoon). This alignment was used
to predict the editing complex of Pol γ by docking of the frayed primer template (primer strand
in chocolate, template strand in orange) onto the apoholoenzyme structure, which we display in
Figure 25B on page 78.

94

BIBLIOGRAPHY

95

BIBLIOGRAPHY

1.
Kaguni, L.S. (2004) DNA polymerase gamma, the mitochondrial replicase. Ann Rev
Biochem, 73, 293-320.
2.
Graziewicz, M.A., Longley, M.J. and Copeland, W.C. (2006) DNA polymerase gamma
in mitochondrial DNA replication and repair. Chem Rev, 106, 383-405.
3.
Yakubovskaya, E., Chen, Z., Carrodeguas, J.A., Kisker, C. and Bogenhagen, D.F. (2006)
Functional human mitochondrial DNA polymerase gamma forms a heterotrimer. J Biol Chem,
281, 374-382.
4.
Longley, M.J., Prasad, R., Srivastava, D.K., Wilson, S.H. and Copeland, W.C. (1998)
Identification of 5'-deoxyribose phosphate lyase activity in human DNA polymerase gamma and
its role in mitochondrial base excision repair in vitro. Proc Natl Acad Sci U S A, 95, 1224412248.
5.
Fan, L., Kim, S., Farr, C.L., Schaefer, K.T., Randolph, K.M., Tainer, J.A. and Kaguni,
L.S. (2006) A novel processive mechanism for DNA synthesis revealed by structure, modeling
and mutagenesis of the accessory subunit of human mitochondrial DNA polymerase. J Mol Biol,
358, 1229-1243.
6.
Lim, S.E., Longley, M.J. and Copeland, W.C. (1999) The mitochondrial p55 accessory
subunit of human DNA polymerase gamma enhances DNA binding, promotes processive DNA
synthesis, and confers N-ethylmaleimide resistance. J Biol Chem, 274, 38197-38203.
7.
Lee, Y.S., Kennedy, W.D. and Yin, Y.W. (2009) Structural insight into processive human
mitochondrial DNA synthesis and disease-related polymerase mutations. Cell, 139, 312-324.
8.
Suomalainen, A., Majander, A., Haltia, M., Somer, H., Lonnqvist, J., Savontaus, M.L.
and Peltonen, L. (1992) Multiple deletions of mitochondrial DNA in several tissues of a patient
with severe retarded depression and familial progressive external ophthalmoplegia. J Clin Invest,
90, 61-66.
9.
Zeviani, M., Servidei, S., Gellera, C., Bertini, E., DiMauro, S. and DiDonato, S. (1989)
An autosomal dominant disorder with multiple deletions of mitochondrial DNA starting at the Dloop region. Nature, 339, 309-311.
10.
Hakonen, A.H., Heiskanen, S., Juvonen, V., Lappalainen, I., Luoma, P.T., Rantamaki,
M., Goethem, G.V., Lofgren, A., Hackman, P., Paetau, A. et al. (2005) Mitochondrial DNA
polymerase W748S mutation: a common cause of autosomal recessive ataxia with ancient
European origin. Am J Hum Genet, 77, 430-441.
11.
Van Goethem, G., Luoma, P., Rantamaki, M., Al Memar, A., Kaakkola, S., Hackman, P.,
Krahe, R., Lofgren, A., Martin, J.J., De Jonghe, P. et al. (2004) POLG mutations in
neurodegenerative disorders with ataxia but no muscle involvement. Neurology, 63, 1251-1257.

96

12.
Winterthun, S., Ferrari, G., He, L., Taylor, R.W., Zeviani, M., Turnbull, D.M., Engelsen,
B.A., Moen, G. and Bindoff, L.A. (2005) Autosomal recessive mitochondrial ataxic syndrome
due to mitochondrial polymerase gamma mutations. Neurology, 64, 1204-1208.
13.
Naviaux, R.K. and Nguyen, K.V. (2004) POLG mutations associated with Alpers'
syndrome and mitochondrial DNA depletion. Ann Neurol, 55, 706-712.
14.
Nguyen, K.V., Ostergaard, E., Ravn, S.H., Balslev, T., Danielsen, E.R., Vardag, A.,
McKiernan, P.J., Gray, G. and Naviaux, R.K. (2005) POLG mutations in Alpers syndrome.
Neurology, 65, 1493-1495.
15.
Suomalainen, A. and Isohanni, P. (2010) Mitochondrial DNA depletion syndromes-many genes, common mechanisms. Neuromuscul Disord, 20, 429-437.
16.
Wong, L.J., Naviaux, R.K., Brunetti-Pierri, N., Zhang, Q., Schmitt, E.S., Truong, C.,
Milone, M., Cohen, B.H., Wical, B., Ganesh, J. et al. (2008) Molecular and clinical genetics of
mitochondrial diseases due to POLG mutations. Hum Mutat, 29, E150-172.
17.
Stumpf, J.D. and Copeland, W.C. (2010) Mitochondrial DNA replication and disease:
insights from DNA polymerase gamma mutations. Cell Mol Life Sci.
18.
Atanassova, N., Fuste, J.M., Wanrooij, S., Macao, B., Goffart, S., Backstrom, S., Farge,
G., Khvorostov, I., Larsson, N.G., Spelbrink, J.N. et al. (2011) Sequence-specific stalling of
DNA polymerase {gamma} and the effects of mutations causing progressive ophthalmoplegia.
Hum Mol Genet, 20, 1212-1223.
19.
Chan, S.S., Longley, M.J. and Copeland, W.C. (2005) The common A467T mutation in
the human mitochondrial DNA polymerase (POLG) compromises catalytic efficiency and
interaction with the accessory subunit. J Biol Chem, 280, 31341-31346.
20.
Kasiviswanathan, R., Longley, M.J., Chan, S.S. and Copeland, W.C. (2009) Disease
mutations in the human mitochondrial DNA polymerase thumb subdomain impart severe defects
in mitochondrial DNA replication. J Biol Chem, 284, 19501-19510.
21.
Luoma, P.T., Luo, N., Loscher, W.N., Farr, C.L., Horvath, R., Wanschitz, J., Kiechl, S.,
Kaguni, L.S. and Suomalainen, A. (2005) Functional defects due to spacer-region mutations of
human mitochondrial DNA polymerase in a family with an ataxia-myopathy syndrome. Hum
Mol Genet, 14, 1907-1920.
22.
Szczepanowska, K. and Foury, F. (2010) A cluster of pathogenic mutations in the 3'-5'
exonuclease domain of DNA polymerase gamma defines a novel module coupling DNA
synthesis and degradation. Hum Mol Genet, 19, 3516-3529.
23.
Chan, S.S. and Copeland, W.C. (2009) DNA polymerase gamma and mitochondrial
disease: understanding the consequence of POLG mutations. Biochim Biophys Acta, 1787, 312319
24.
Brieba, L.G., Eichman, B.F., Kokoska, R.J., Doublie, S., Kunkel, T.A. and Ellenberger,
T. (2004) Structural basis for the dual coding potential of 8-oxoguanosine by a high-fidelity
DNA polymerase. EMBO J, 23, 3452-3461
25.
Doublie, S., Tabor, S., Long, A.M., Richardson, C.C. and Ellenberger, T. (1998) Crystal
structure of a bacteriophage T7 DNA replication complex at 2.2 A resolution. Nature, 391, 251258.
97

26.
Li, Y., Korolev, S. and Waksman, G. (1998) Crystal structures of open and closed forms
of binary and ternary complexes of the large fragment of Thermus aquaticus DNA polymerase I:
structural basis for nucleotide incorporation. EMBO J, 17, 7514-7525.
27.

Berdis, A.J. (2009) Mechanisms of DNA polymerases. Chem Rev, 109, 2862-2879.

28.
Johnson, K.A. (2010) The kinetic and chemical mechanism of high-fidelity DNA
polymerases. Biochim Biophys Acta, 1804, 1041-1048.
29.
Lee, H.R., Helquist, S.A., Kool, E.T. and Johnson, K.A. (2008) Base pair hydrogen
bonds are essential for proofreading selectivity by the human mitochondrial DNA polymerase. J
Biol Chem, 283, 14411-14416.
30.
Lee, H.R., Helquist, S.A., Kool, E.T. and Johnson, K.A. (2008) Importance of hydrogen
bonding for efficiency and specificity of the human mitochondrial DNA polymerase. J Biol
Chem, 283, 14402-14410.
31.
Polesky, A.H., Dahlberg, M.E., Benkovic, S.J., Grindley, N.D. and Joyce, C.M. (1992)
Side chains involved in catalysis of the polymerase reaction of DNA polymerase I from
Escherichia coli. J Biol Chem, 267, 8417-8428.
32.
Polesky, A.H., Steitz, T.A., Grindley, N.D. and Joyce, C.M. (1990) Identification of
residues critical for the polymerase activity of the Klenow fragment of DNA polymerase I from
Escherichia coli. J Biol Chem, 265, 14579-14591.
33.
Graziewicz, M.A., Longley, M.J., Bienstock, R.J., Zeviani, M. and Copeland, W.C.
(2004) Structure-function defects of human mitochondrial DNA polymerase in autosomal
dominant progressive external ophthalmoplegia. Nat Struct Mol Biol, 11, 770-776.
34.
Van Goethem, G., Dermaut, B., Lofgren, A., Martin, J.J. and Van Broeckhoven, C.
(2001) Mutation of POLG is associated with progressive external ophthalmoplegia characterized
by mtDNA deletions. Nat Genet, 28, 211-212.
35.
Johnson, S.J., Taylor, J.S. and Beese, L.S. (2003) Processive DNA synthesis observed in
a polymerase crystal suggests a mechanism for the prevention of frameshift mutations. Proc Natl
Acad Sci U S A, 100, 3895-3900.
36.
Jin, Z. and Johnson, K.A. Role of a GAG hinge in the nucleotide-induced conformational
change governing nucleotide specificity by T7 DNA polymerase. J Biol Chem, 286, 1312-1322.
37.
Baruffini, E., Horvath, R., Dallabona, C., Czermin, B., Lamantea, E., Bindoff, L.,
Invernizzi, F., Ferrero, I., Zeviani, M. and Lodi, T. (2011) Predicting the contribution of novel
POLG mutations to human disease through analysis in yeast model. Mitochondrion, 11, 182-190.
38.
Luo, N. and Kaguni, L.S. (2005) Mutations in the spacer region of Drosophila
mitochondrial DNA polymerase affect DNA binding, processivity, and the balance between Pol
and Exo function. J Biol Chem, 280, 2491-2497.
39.
Murakami, K.S., Davydova, E.K. and Rothman-Denes, L.B. (2008) X-ray crystal
structure of the polymerase domain of the bacteriophage N4 virion RNA polymerase. Proc Natl
Acad Sci U S A, 105, 5046-5051.
40.
Jeruzalmi, D. and Steitz, T.A. (1998) Structure of T7 RNA polymerase complexed to the
transcriptional inhibitor T7 lysozyme. EMBO J, 17, 4101-4113.
98

41.
Gleghorn, M.L., Davydova, E.K., Rothman-Denes, L.B. and Murakami, K.S. (2008)
Structural basis for DNA-hairpin promoter recognition by the bacteriophage N4 virion RNA
polymerase. Mol Cell, 32, 707-717.
42.
Baruffini, E., Ferrero, I. and Foury, F. (2007) Mitochondrial DNA defects in
Saccharomyces cerevisiae caused by functional interactions between DNA polymerase gamma
mutations associated with disease in human. Biochim Biophys Acta, 1772, 1225-1235.
43.
Stumpf, J.D., Bailey, C.M., Spell, D., Stillwagon, M., Anderson, K.S. and Copeland,
W.C. (2010) mip1 containing mutations associated with mitochondrial disease causes
mutagenesis and depletion of mtDNA in Saccharomyces cerevisiae. Hum Mol Genet, 19, 21232133.
44.
Baruffini, E. and Lodi, T. Construction and validation of a yeast model system for
studying in vivo the susceptibility to nucleoside analogues of DNA polymerase gamma allelic
variants. Mitochondrion, 10, 183-187.
45.
Lee, Y.S., Lee, S., Demeler, B., Molineux, I.J., Johnson, K.A. and Yin, Y.W. (2010)
Each monomer of the dimeric accessory protein for human mitochondrial DNA polymerase has a
distinct role in conferring processivity. J Biol Chem, 285, 1490-1499.
46.
Lee, Y.S., Johnson, K.A., Molineux, I.J. and Yin, Y.W. (2010) A single mutation in
human mitochondrial DNA polymerase Pol gammaA affects both polymerization and
proofreading activities of only the holoenzyme. J Biol Chem, 285, 28105-28116.
47.
Chan, S.S., Longley, M.J. and Copeland, W.C. (2006) Modulation of the W748S
mutation in DNA polymerase gamma by the E1143G polymorphismin mitochondrial disorders.
Hum Mol Genet, 15, 3473-3483.
48.
Palin, E.J., Lesonen, A., Farr, C.L., Euro, L., Suomalainen, A. and Kaguni, L.S.
Functional analysis of H. sapiens DNA polymerase gamma spacer mutation W748S with and
without common variant E1143G. Biochim Biophys Acta, 1802, 545-551.
49.
Naviaux, R.K. and Nguyen, K.V. (2005) POLG mutations associated with Alpers
syndrome and mitochondrial DNA depletion. Ann Neurol, 58, 491.

99

CHAPTER 3: Mapping 136 Pathogenic Mutations into Functional Modules in Human
DNA Polymerase γ Establishes Predictive Genotype-phenotype Correlations for the
Complete Spectrum of POLG Syndromes.
Gregory A. Farnum1, Anssi Nurminen2, and Laurie S. Kaguni1,2

1

Department of Biochemistry and Molecular Biology and Center for Mitochondrial Science and

Medicine, Michigan State University, East Lansing, MI 48824-1319, USA; 2Institute of
Biomedical Technology, University of Tampere, 33014 Tampere, Finland;

100

ABSTRACT
We establish for the first time genotype-phenotype correlations for the complete spectrum
of POLG syndromes, by refining our previously described protocol for mapping pathogenic
mutations in the human POLG1 gene to functional clusters in the catalytic core of the
mitochondrial replicase, Pol γ (1). We assigned 136 mutations to five clusters and identify
segments of primary sequence that can be used to delimit the boundaries of each cluster. We report
that compound heterozygotes with two mutations from different clusters manifested more severe,
earlier onset POLG syndromes, whereas two mutations from the same cluster are less common
and generally are associated with less severe, later onset POLG syndromes. We also show that
specific cluster combinations are more severe than others, and have a higher likelihood to manifest
at an earlier age. Our clustering method provides a powerful tool to predict the pathogenic potential
and predicted disease phenotype of novel variants and mutations in POLG1, the most common
nuclear gene underlying mitochondrial disorders. We propose that such a prediction tool would be
useful for routine diagnostics for mitochondrial disorders.

101

INTRODUCTION
Mitochondrial dysfunction due to impaired energy production via oxidative
phosphorylation (OXPHOS) causes a variety of diseases, known collectively as mitochondrial
disorders (2). Base substitutions, deletions or depletion of mitochondrial DNA (mtDNA) resulting
in production of dysfunctional and/ or depletion of OXPHOS proteins is an important cause of
mitochondrial dysfunction (2). In animal mitochondria, Pol γ is the only known DNA polymerase,
and therefore it is responsible for maintenance of mtDNA integrity associated with mtDNA
replication and repair (3). Human Pol γ is a heterotrimer consisting of a catalytic subunit, Pol γA,
and a dimer of an accessory subunit, Pol γB (4). Encoded by the POLG1 gene, Pol γA, known as
the catalytic core, is a 140 kDa polypeptide that contains the 5'-3' DNA polymerase (pol), 3'-5'
exonuclease (exo), and 5'-dRP lyase activities (3).

Mutations in the POLG1 gene lead to

accumulation of mtDNA deletions as well as mtDNA depletion, manifesting mitochondrial
disorders termed POLG syndromes (5). The most severe form of POLG syndrome, known as
Alpers syndrome, is caused by compound heterozygosity of two recessive POLG mutations, and
leads to hepatocerebral mtDNA depletion syndrome during infancy and death at an early age (6).
Late onset POLG syndromes, such as progressive external ophthalmoplegia (PEO) can be caused
by dominant or recessive compound heterozygous mutations, and are associated with varying
degrees of tissue specific mtDNA depletion and/ or single or multiple mtDNA deletions (7).
The crystal structure of Pol γ was determined in its apoenzyme form (PDB code 3IKM,
(8)). Its catalytic core has three major domains: an N-terminal Exo domain that contains the exo
active site; a C-terminal Pol domain that contains the pol active site; and a spacer domain that
separates the Exo and Pol domains in primary sequence. Three subdomains are defined in the Pol
domain: the palm subdomain represents the most highly conserved structural module between non-

102

homologous DNA polymerases, and contains the pol catalytic site coordinating two Mg2+ ions;
the fingers subdomain is involved in binding the incoming dNTP substrate, and the thumb
subdomain forms a major surface of the DNA binding channel. The spacer domain comprises two
subdomains: the accessory interacting domain (AID), which forms the major hydrophobic contact
with the proximal accessory subunit, and the intrinsic processivity (IP) domain, which forms a
region of the upstream DNA binding channel and does not contact the accessory subunits. Both
subdomains form distinct regions of the upstream DNA binding channel, but the flexible AID
contributes to DNA binding affinity only when the accessory subunit stabilizes it in the correct
position (9). The extreme N-terminal residues of the catalytic core do not show sequence
conservation with other Exo domains in the family A polymerase group and are thus designated
as a separate domain N-terminal domain (NTD), which presumably contains the mitochondrial
leader sequence.
Pol γ is a member of the family A polymerase group and shares the highest sequence
similarity with bacteriophage T7 DNA polymerase (T7 Pol) (8). We generated a structural model
of the Pol γ ternary complex by superimposition of the apo-holoenzyme Pol γ structure (PDB code
3IKM) (1) with the structure of T7 DNA polymerase (PDB code 1T8E) with bound primer
template DNA, incoming ddNTP and Mg2+ ions. We then mapped 58 Alpers mutations onto this
structural model and reported that they cluster into distinct regions, which we termed clusters 1-5
(1). Because Pol γ is a family A polymerase, functional insight can be extrapolated from the
extensive studies of conserved elements in bacterial DNA polymerase I (Pol I) and T7 Pol (1015). This and evaluation of the unique features of Pol γ by biochemical analysis of yeast, fly, and
human Pol γ variants (16-18), revealed that each cluster defines a unique functional region, which
exhibits a distinct biochemical defect when affected by a pathogenic mutation (1). Cluster 1
103

mutations are predicted to cause a primary defect in pol activity, and affect residues involved
directly in catalysis, or indirectly by affecting architectural residues that may disrupt the position
of catalytic residues. Catalytic residues include those binding the two Mg2+ ions, those that make
contact with the incoming dNTP, and those that make contact with the first, second and third
nucleotide pairs in the nascent dsDNA, which are critical for correct positioning of the substrate
in the pol active site. For example, the human pathogenic mutation R943H affects an amino acid
residue that contacts the γ phosphate of the incoming dNTP, and has been shown in vitro to have
a 30-fold increase in Km for dNTPs and a 5-fold decrease in kcat, which together reduce pol activity
150-fold without affecting DNA binding affinity (19). Cluster 2 mutations are predicted to cause
a primary defect in DNA binding affinity, and affect residues of the IP and AID that form the
upstream DNA binding channel, which enhance DNA binding affinity via contacts with
nucleotides upstream of the third base pair of primer template.

Biochemical analysis of

recombinant fly Pol γ with a triple alanine substitution in amino acid residues K768/D769/F770
are located in the DNA binding channel wall of the IP domain, and cause a small decrease in
binding affinity (17). Cluster 3 mutations are predicted to affect the pol: exo activity ratio and
may have variable defects in DNA binding affinity. Many of the cluster 3 mutations map to a
helix-coil-helix module (residues 295-312) located in the Exo domain that has been termed the
"orienter" module (18). L304R of the "orienter" module has been studied biochemically in
recombinant yeast Pol γ, and exhibits 3-fold increased exo activity, 20-fold decreased pol activity
and 10-fold decreased DNA binding affinity (18). In addition, cluster 3 mutations map to the
partitioning loop, which is a novel module conserved in Pol γ (residues 1050-1095) that is located
between the fingers and palm subdomains and is not present in any other known polymerase (1).
To date, no biochemical data is available for residues of the partitioning loop, although it has been
104

shown that yeast strains homozygous for a mutation that is equivalent to G1051R in human Pol γ
cause a 10-fold increase in point mutational frequency in vivo (20). Cluster 4 mutations map to
the Exo domain along the distal accessory subunit interface, and are predicted to cause a
biochemical defect similar to the R232G variant, which was shown in vitro to have reduced pol
rate, increased exo activity and wild type (WT) DNA binding affinity (21). Cluster 5 mutations
are located in the periphery of the IP subdomain and have not been shown to cause a biochemical
defect. For example, human recombinant Pol γ harboring R627Q or R627W mutations exhibited
no defects in vitro when analyzed for DNA binding affinity, pol activity and stimulation by the
accessory subunit (22).
In applying our clustering model to the reported Alpers mutations, we found that Alpers
syndrome was found only in compound heterozygous patients bearing two mutations from
different clusters in Pol γ (1). Since publication, several new reports have been published that
have added dozens of new mutations and mutation combinations to the POLG syndrome library
(23-26). Here, we examine potential genotype-phenotype correlations by analyzing all reported
POLG mutations available to date, not restricted to a phenotype or age-of-onset, utilizing our
clustering model. In total, we assign 136 mutations to the five clusters and identify segments of
primary sequence that can be used to delimit the boundaries of each cluster. We demonstrate the
validity of our clustering model as the first to establish genotype-phenotype correlations for POLG
syndromes, and show that it can be used to predict the pathogenic potential and biochemical defects
of novel mutations, and to provide information about the likely severity of the POLG syndrome
for compound heterozygotes with novel combinations of mutations.

105

METHODS
Computational analysis
We docked primer-template DNA (ptDNA) into the putative DNA-binding channel of the
Pol γ apo-holoenzyme crystal structure (PDB code 3IKM, (8)) by superposition of the closed
ternary complex of T7 Pol bound to ptDNA and dNTP (PDB code 1T8E, (27)) using PyMOL
(http://www.pymol.org/). As we described previously, we found that the best overall alignment
was obtained by aligning the palm subdomain of Pol γA (residues 815-910, 1095-1239) to the
palm subdomain of T7 Pol (residues 409-487, 611-704) (1). This model was used to map and
evaluate all reported POLG disease mutations.
Statistical analysis
We used Pearson’s chi-square analysis to calculate the significance of each age of onset
(infantile, childhood, juvenile, adult), and the mutation combination affecting the same (=0) or
different cluster (=1) of Pol γ. Logistic regression analysis was used to calculate odds ratios for
having an infantile-onset disease. We conducted all statistical analyses with Stata (version 11.0).
Two-sided p values were used with a significance level of 0·05.

106

RESULTS
Cluster assignment of all reported POLG disease mutations
The Alpers mutations represent only a subset of the total library of reported diseasecausing mutations in POLG. To investigate further the utility of the five functional clusters in Pol
 that we defined earlier to evaluate Alpers mutations (1), we compiled a list of all pathogenic
point

mutations

reported

to

date

by

combining

the

POLG

database

entries

(http://tools.niehs.nih.gov/polg/) with new data from recent reports (23-26, 28, 29). The list
comprises a total of 136 pathogenic point mutations, including both dominant and recessive
mutations. We included in our analysis POLG mutations that represented the only disease-causing
defect identified in the patient, exclusive of mutations in other genes associated with mitochondrial
dysfunction, and excluded consideration of null mutations. By excluding patients with mutations
in multiple genes associated with mitochondrial dysfunction, a clear link can be established
between functional defects in Pol γ and the severity of disease manifestations. Furthermore, we
considered a pathogenic mutation to be dominant only if the family history demonstrated a
dominant mode of inheritance, leaving out potential de novo mutations.
We mapped each of the pathogenic mutations onto our structural model and assigned them
to a cluster by evaluating them individually. Although a cluster may contain residues that are
distant in primary sequence, assignment was straightforward in most cases because each cluster
occupies a distinct structural region in the Pol γ tertiary structure. Some mutations mapped to
areas equidistant from two different clusters, and because we did not define explicitly the
boundaries of the structural regions for each cluster in our previous analysis, we made use of
additional criteria for cluster assignment. To do so, we analyzed each point mutation based on
reported biochemical data to predict the functional defect it would cause. In this way, cluster
assignment is based not only on structural location but also upon functional insight. In total, we
107

assigned 136 pathogenic mutations to the five functional clusters we defined earlier (1) as shown
in Figure 32. Due to the large library of reported mutations, we concluded that a sufficient number
of mutations were assigned to each cluster to use their map positions as the sole means of defining
the structural region and primary sequence occupied by each cluster. Mutations from each cluster
are divided into several subclusters of 10-100 amino acid residues, and are distributed across the
length of the POLG1 gene (Figure 32, upper schematic).

Despite the large separation of

subclusters in primary sequence, the elements of a cluster fold into compact structural regions as
illustrated in Figure 32 (lower panel).
68 mutations were assigned to cluster 1, defining seven subclusters located in the NTD,
Exo and Pol domains. The primary biochemical defect caused by a cluster 1 mutation is predicted
to be reduced pol activity. As expected, most mutations affect residues of the Pol domain,
including the five conserved motifs within the Pol domain of family A polymerases that are
essential for 5'-3' DNA polymerase activity (Figure 33, upper schematic, (12)). Subcluster 1e
(residues 914-966) spans the O-helix and its environs (Figure 33, lower panel), the function of
which is to bind the correct dNTP substrate by transitioning between an open and closed complex
(30). The conserved amino acid residues on the O-helix are known as the Pol B motif (residues
943-958), and function to establish specific contacts with correctly base paired dNTP in the closed
conformation (12). POLG mutations that disrupt the specific contacts with the incoming dNTP
(H932Y, R943H, K947R, Y951N, and Y955C) will reduce fidelity and increase Km (dNTP),
without affecting DNA binding affinity (19, 31). Thus, mutations affecting this site are most likely
dominant because they are capable of competing for dNTP binding with wild type (WT) Pol γ but
are unable to polymerize nucleotides effectively, and are predicted to cause mtDNA damage that
is associated with enzyme stalling (32).
108

Figure 32. Clustering of 136 pathogenic mutations within five functional modules in the catalytic subunit of human Pol γ.

109

Figure 32 (cont’d).

Upper panel, schematic diagram of the human POLG1 gene illustrating the clustering of 136 pathogenic mutations into discrete
blocks of amino acid residues, which we term subclusters. Mutant alleles and subclusters are colored according to the cluster to which
they belong: cluster 1, green; cluster 2, yellow; cluster 3, red; cluster 4, blue; cluster 5, cyan Figure 32 (Cont’d) (see the text for
details). The palm (residues 815-910 and 1096-1239), fingers (residues 911-1048), and thumb (Th, residues 440-475 and 785-814)
subdomains of the Pol domain are colored pink, and the partitioning loop (PL, residues 1049-1095) is red. The accessory (subunit)
interacting domain (AID, residues 476-570) and the intrinsic processivity (IP, residues 571-784) subdomains of the spacer domain are
colored in magenta.

110

Figure 32 (cont’d) The N-terminal domain (NTD, residues 1-170) and the Exo domain (residues
171-439) are colored in purple. Lower panel, structural model of the human Pol γ apoholoenzyme (PDB code 3IKM) with docked primer template DNA shown as orange ribbon (see
Computational methods for details). The catalytic subunit of Pol γ is shown as a cartoon
representation of the secondary structural elements (SSEs), with regions defined by clusters
illustrated as space-filled modules, colored according to the schematic. The proximal and distal
accessory subunits are shown as surface representations in light and dark gray, respectively.
Subcluster 1d (residues 848-895) comprises the RR loop (residues 845-863) (1) and the
conserved Pol A motif (residues 887-896) of the palm subdomain (Figure 33) (3). The RR loop is
equivalent to motif 2 in Pol I, and is critical for binding correctly base-paired ptDNA in the minor
groove and the template DNA backbone (12). The Pol A motif in Pol I contacts the primer strand,
binds Mg2+ via amino acid residue D705 (D890 in Pol ) and discriminates against ribonucleotide
incorporation via residue E710 (E895 in Pol ) (33). Mutations in the Pol A motif at either D890
or E895 in Pol  would likely manifest a dominant lethal phenotype, and the only reported patient
to date with the E895G mutation died immediately after birth (34). Motif 2 mutations should
always cause reduced pol activity but may also cause a DNA binding defect depending on the
residue, such as the 5-fold defect reported for G848S and R852C (35). The DNA binding affinity
of T851A, R853Q, Q879H, T885S may decrease slightly, but the reduction in pol activity is much
greater (35). Overall, we feel it is unlikely for dominant mutations to reside in motif 2 because
these mutant forms of Pol γ would not compete with WT Pol γ for DNA binding.

111

Figure 33. Architectural and functional subclusters of the Pol domain. Upper panel, schematic diagram of the POLG1 gene as
shown in Figure 32, with an additional upper section indicating the location of the three Exo motifs, labeled as I, II, and III, and the six
Pol motifs, labeled as 1 (motif 1), 2 (motif 2), A (Pol A motif), B (Pol B motif), 6 (motif 6) and C (Pol C motif), which are conserved
throughout family A polymerases (see text). Motifs and subclusters that are illustrated in the bottom panels are in shown bold.
Bottom panels, SSEs and transparent surface representation of the palm and fingers subdomains of Pol γ (PDB code 3IKM) are shown
(subdomains are colored in gray in the middle section of the schematic). Bottom-left panel, the five motifs of the Pol domain, which
are
112

Figure 33 (cont’d) colored in pink in the upper section of the schematic, are shown as pink SSEs
and are labeled accordingly. Docked primer template DNA is shown as orange ribbon, Mg2+
ions are shown as orange spheres, and the incoming dNTP is shown as orange sticks (see
Computational methods for details). Bottom-right panel, subclusters 1D, 1E and 1F encompass
one or more of the five conserved motifs of the Pol domain and are colored in light green.
Subclusters 1A, 1B, 1C and 1G, colored in dark green, are located further structurally from the
pol active site, and are considered to play an architectural role.
Subcluster 1f (residues 1104-1138) comprises motif 6 (residues 1097-1110) and the Pol C
motif (residues 1134-1141) in the palm subdomain (Figure 33) (according to the nomenclature
defined for Pol I (12)). Motif 6 is located on the Q-helix and binds correctly base paired template
strand and the ptDNA minor groove in Pol I via the residues N845 and Q849 ((13), equivalent to
N1098 and Q1102 in Pol respectively. The Pol C motif in Pol  binds Mg2+ via D1135, whereas
H1134 and E1136 contact the primer strand. Subcluster 1g (residues 1157-1196) maps to a Cterminal region of the palm subdomain and forms an anti-parallel beta strand adjacent to the Pol
A motif. Only K1191 is predicted to contact the primer terminus, whereas the rest of subcluster
1g serves an architectural role (1). Mutations affecting motif 6 have been reported to cause DNA
binding defects along with pol defects in Pol I Klenow (13). However, an alanine substitution at
the equivalent residue to Pol  H1134 in Klenow (H881A) caused a decrease in kcat from 6- to 66fold (13) but retained DNA binding affinity, suggesting that a mutation in H1134 might be
dominant.
We assigned mutations to subclusters 1d-g with high confidence because there is extensive
biochemical data available for the Pol domain, in addition to the presence of the highly conserved
amino acid sequence motifs. In contrast, a subcluster of seven mutations (R417T, C418R, G426S,
L428P, M430L, G431V, S433C) mapped to the G-helix (Figure 33, lower right panel), a structural
element that had not been studied previously. According to the current Exo domain assignment
of residues 170-440 by Lee et al. (8), this subcluster would reside in the Exo domain, although
113

earlier T7 Pol (15) and Pol I structures (36) label the G-helix as part of the palm subdomain within
the Pol domain. Regardless of the domain assignment, this subcluster is adjacent to motif 2 and
motif 6, and we propose it serves an architectural role that contributes to pol activity indirectly,
and assign it as subcluster 1c. Four mutations from the NTD were also observed to be closer
structurally to functional Pol motifs as compared to Exo motifs; we thus assigned D136E and
A143V as subcluster 1b (residues 136-143), and both L83P and F88L were assigned as subcluster
1a (residues 83-88).
25 mutations from cluster 2, which we predict to cause primary defects in DNA binding
affinity, outline the putative DNA binding channel, and map to four subclusters within the thumb
subdomain of the Pol domain (2a, residues 463-468), and the AID (2b, residues 497-517) and IP
(2c, residues 561-617 and 2d, 752-767) subdomains of the spacer domain. Subcluster 2c contains
motif 1, which in Pol I was shown to fold into a loop and binds DNA in the channel (12). Motif
1, together with subcluster 2d, form the major face of the DNA binding channel. Subcluster 2a
maps to a region of the thumb subdomain at the accessory subunit interface where A467T, N468D,
and L463F are positioned. Our group has characterized residue A467T biochemically, and it was
shown to retain 70% of WT DNA binding affinity in a reconstituted holoenzyme form (22).
Extensive biochemical analysis of the spacer domain in general has demonstrated that the
hydrophobic core of the IP subdomain is critical for shaping the DNA binding channel wall (17,
22). Mutations that alter the IP subdomain, such as A467T, are predicted to perturb the channel
to block DNA from entering, which is then observed as reduced DNA binding affinity.
8 mutations from cluster 3 map to the orienter module (18) of the Exo domain, and another
12 mutations from the same cluster map to the partitioning loop (1) of the Pol domain, which are
defined by two subclusters (3b, residues 303-319 and 3d, 1047-1096, respectively). Additionally,

114

a recent report from the Foury group (37) has shown that variants of yeast Pol γ bearing
substitutions in residues of the thumb subdomain M602I (R802 in Hs) and the Exo II motif of the
Exo domain A228V, R231K, R233W (S272, R275, H277 in Hs) also produce the altered pol:exo
activity ratio that characterizes cluster 3 mutations. Therefore, we assign pathogenic residues
R807P, R807C, R807H, and R804T as subcluster 3c (residues 804-807), and the Exo II motif
mutations G268A, R275Q, and H277L as subcluster 3a (residues 268-277). In support of the
proposed subcluster 3c, we note that the SYW triple alanine substitution in recombinant fly Pol γ,
which was found to exhibit increased exo activity with both decreased pol activity and DNA
binding affinity (17), maps adjacent to subcluster 3c in human Pol γ (residues S799/F800/W801).
Similarly, the Exo II motif has been demonstrated to alter the pol: exo activity ratio in biochemical
variants of Pol I by decreasing the affinity of the primer strand for the exo active site, consistent
with the decrease in exo activity and WT pol activity observed for the R233W variant in yeast Pol
γ (H277 in human Pol γ) (37).
Six mutations from cluster 4 C224Y, R227P, R227W, R232G, R232H and L244P map to
a single region on the subunit interface of the Exo domain (subcluster 4a, residues 224-244).
Biochemical studies of a human R232G variant in reconstituted holoenzyme form showed a
decrease in pol rate and an increase in exo activity, with unchanged DNA binding affinity that
derive from loss of a direct stimulation of pol activity by the distal accessory subunit (21). We
propose that the other residues of cluster 4 have similar biochemical characteristics.
Ten mutations from cluster 5 are also located in the IP domain, but they map to the distal
region far removed from the DNA binding channel, and define subcluster 5a (residues 623-648)
and subcluster 5b (residues 737-749).

115

Cluster assignment can be used to predict the pathogenicity of a novel mutation.
Our clustering model provides an annotated guide to assign function to the entire
mutational spectrum in the POLG1 gene, and we propose that it can be used to evaluate the
potential pathogenicity of novel POLG mutations. In particular, when a novel mutation maps to a
specific subcluster within the five functional clusters we have defined, we predict that it will likely
be pathogenic and result in the biochemical defect associated with that subcluster. At the same
time, we would argue that mutations that do not map to these specific amino acid sequence blocks
are more likely to be neutral polymorphisms. In support of this model, we found that only 12 out
of 87 reported single nucleotide polymorphisms (SNPs) from the human dbSNP database
(http://www.ncbi.nlm.nih.gov/projects/SNP/) map within subclusters (Figure 34). Three of these
are conservative substitutions of highly similar amino acids (F1092L, F1164L and F1164I), and
likely have no functional consequences. These examples would suggest that although a mutation
may affect a residue within a functional subcluster, conservative substitutions with amino acids
with highly similar chemical properties may not be pathogenic, and this possibility should be taken
into consideration when evaluating novel mutations. Additionally, missense mutations reported as
SNPs in the SNP database are not necessarily non-pathogenic, as they could be uncommon
recessive mutations that are masked when paired with a wild type allele.
Although the mutation library is large, the POLG1 mutational map is likely not yet
saturated, so routine sequencing of patients with POLG syndrome will continue to fill out the
spectrum. We predict that most novel mutations will map within our current clusters because they
are by definition "mutational hotspots" and importantly, we observe from their positions within
the crystal structure that they encompass the known functional motifs required for the biochemical
activities of Pol γ. For example, based upon the high sequence conservation with family A
polymerases, we note that the essential elements of the pol active site such as the Pol A-C motifs,
116

motif 2, and motif 6 are each populated by human pathogenic mutations that map within our
current cluster 1 (Figures 32 and 33).

117

Figure 34. SNPs reported in the POLG1 gene rarely map to clusters. Of 87 reported SNPs in the dbSNP database, 75 (86%, black
boxes) map outside the defined pathogenic clusters (shown as colored regions in the schematic). The remaining 12 SNPs (magenta
boxes) that map within the clusters can be considered as non-deleterious changes within a high-risk region of the POLG tertiary
structure. However, in our view, the possible pathogenicity of the SNPs reported in cluster 4 (dark blue) warrants careful experimental
reevaluation.
118

Cluster combinations of compound heterozygotes correlate with age of onset of POLG
syndrome.
Previously, we evaluated the combination of mutations in compound heterozygotes that
cause Alpers syndrome using our original clustering model, and found that Alpers was triggered
exclusively by the combination of two mutations from different clusters (1). Here, we evaluate all
mutations causing POLG syndromes (caused by compound heterozygosity) using the model
refined by the designation of subclusters within the original five functional clusters. Whereas
clinical presentation and age of onset are extremely variable for POLG syndromes, in general the
severity of symptoms decreases with increasing age of onset (38). We were able to catalogue a
total of 341 compound heterozygous patients harboring two pathogenic mutations and we
classified each patient by the age of onset reported. Each combination was assigned to one of four
age groups: infantile, < 3 years of age; childhood, ages 3-12; juvenile, 13-20 years; and adult, >
20 years. The data are compiled and evaluated in Figure 35. Compound heterozygotes with two
mutations from the same cluster were less common (113/341) and were generally associated with
late onset POLG syndromes (Figure 35, upper panel). Compound heterozygotes with two
mutations from different clusters were much more common (228/341), and comprised 109/119 of
the infantile onset combinations. Clearly, the combination of two different clusters is not just a
specific trigger for Alpers syndrome; rather, it establishes a universal genotype-phenotype
correlation for all POLG syndromes.

119

Figure 35. Analysis of mutation combinations as cluster combinations reveals predictive
genotype-phenotype correlations. The data used to compile the information in this figure was
120

Figure 35 (cont’d) derived from the literature listed under supplemental references in appendix
C. The number of mutation combinations manifesting POLG syndrome at each of the four age
groups is shown in each panel for specific cluster combinations. Age of onset trends can be used
to predict the severity of POLG syndrome for an individual with compound heterozygous
mutations in POLG1. Upper panel, POLG syndrome age of onset trends for mutation
combinations of two mutations from different clusters versus the same cluster. Middle panel,
patient data show trends in which severe cluster combinations have an earlier age of onset,
whereas less severe or uncommon cluster combinations have a later age of onset. Lower panel,
earlier age of onset trends for more severe cluster/ subcluster combinations.
Critical functions: clusters 1, 3, and 4
We predict that mutations affecting cluster 1 will reduce primarily pol activity, and these
have been associated with increased replication fork stalling in vivo (32). The stalling phenotype
is characterized by slow progression of or suspended replication forks, and causes an accumulation
of replication intermediates that can lead to double-stranded DNA breaks, mtDNA deletions and
base substitution mutations (39-41). It is thus unlikely that a compound heterozygote with cluster
1 mutations would be viable unless the defects are very mild. Indeed, only nine of these
combinations have been reported (Figure 36), and involve mutations such as G923D and A957S
that retain 25% WT pol (19), which is a relatively mild pol defect compared to other cluster 1
mutations, such as the 100-fold reduction of pol activity engendered by the Y955C mutation (32).
Although the partitioning loop has not been studied biochemically, cluster 3 mutations may
produce a biochemical phenotype similar to that of the L260R variant in yeast Pol γ (L304R in Hs
Pol γ), which is characterized by an increase in exo activity, and decreases in pol activity and DNA
binding affinity (18). Only four combinations have been reported for compound heterozygotes
with two cluster 3 mutations (Figure 36), which may indicate that most of these combinations are
lethal. Cluster 4 mutations are predicted to increase exo activity and decrease pol rate, while
retaining WT DNA binding affinity (21). Interestingly, cluster 4 mutations have not been reported
in combination with other cluster 4 mutations, or together with cluster 3 mutations. When Pol γ is
stalled, the primer strand can translocate between the exo active site the pol active site, a mode
121

described as idling by Atanassova et al. (32). We speculate that cluster 3 and 4 mutants may
manifest an idling phenotype because of their enhanced exo: pol activity ratios, which is likely to
increase non-productive turnover by mutant Pol γ, via exonucleolytic hydrolysis of correctly
incorporated nucleotides. In compound heterozygotes, two idling Pol γs may be lethal, explaining
the absence of cluster 3 + 4 combinations, as well as cluster 4 + 4 combinations. Taken together,
the lack of complementation observed for clusters 1, 3, and 4 suggests that they serve essential
functions in mtDNA metabolism.

Figure 36. Analysis of all compound heterozygous mutation combinations as cluster
combinations. The data used to compile the information in this figure was derived from the
literature listed under supplemental references in appendix C. The number of mutation
combinations manifesting POLG syndrome at each of the four age groups is shown for all cluster
combinations for which patient data have been reported; no patient data have been reported for
the three cluster combinations 3+4, 4+4 and 4+5. Age of onset trends can be used to predict the
severity of POLG syndrome for an individual with compound heterozygous mutations in
POLG1. Patient data show trends in which severe cluster combinations have an earlier age of
onset, whereas less severe or uncommon cluster combinations have a later age of onset.

122

Moderately severe dysfunction: clusters 2 and 5
Mutations affecting cluster 2 will most likely cause a primary defect in DNA binding
affinity. Processivity, defined as the number of nucleotides incorporated in a single DNA binding
event, varies directly with pol rate and DNA binding affinity and as a consequence, cluster 2
mutations reduce processivity. Individuals carrying homozygous A467T mutations present as
juveniles with POLG syndrome (26, 42, 43). As we showed that Pol γ carrying the A467T
substitution shows a DNA binding affinity 70% of that of WT (22), this suggests that moderate
reductions in DNA binding and pol activity may be tolerated throughout childhood. It thus seems
likely that most same cluster combinations of cluster 2 mutations will not be lethal, and instead
will lead to development of later-onset POLG syndrome. Consistent with this hypothesis, 42 such
combinations have been reported, and the majority show juvenile or adult onset (Figure 35, middle
panel).
W748S, R627Q and R627W are cluster 5 mutations and have been studied in vitro in
reconstituted human Pol γ holoenzyme, but all three revealed no biochemical phenotype (22, 44).
These mutations map to the distal surface of the IP domain, far removed from the DNA binding
channel. Despite the lack of a documented biochemical phenotype, cluster 5 mutations have been
reported in 140 compound heterozygotes with POLG syndrome manifesting at all ages (Figures
35 and 36). Therefore, this region clearly carries a biologically-relevant function in mtDNA
metabolism because cluster 5 mutations cause mtDNA instability (44, 45). As for cluster 2
mutations, cluster 5 mutations are found together in compound heterozygote individuals with
juvenile or adult onset POLG syndrome. Compound heterozygotes with a cluster 2 mutation and
a cluster 5 mutation have been reported in 40 patients, and a majority of the combinations resulted
in juvenile or adult onset POLG syndrome (Figure 35, middle panel). Notably, this cluster
combination is the only one that did not typically cause infantile or childhood onset POLG
123

syndromes. To this point, 39/39 patients that were homozygous for W748S only developed a mild
adult-onset POLG syndrome (46-48). In sum, cluster 2 and cluster 5 mutations, in contrast to those
in clusters 1, 3 and 4, are observed to complement each other, suggesting they serve less critical
functions in mtDNA metabolism.

Severe combination: cluster 1 + 2
Compound heterozygotes with cluster 1 and cluster 2 mutations have been reported in 70
patients, and are the most abundant of all cluster combinations (Figure 35, middle and lower
panels). The majority (44/70) of these combinations caused infantile onset POLG syndrome. In
fact, the most severe combinations observed in this study are the subcluster combinations 1D + 2A
and 1E + 2A. These combinations cause infantile onset POLG syndrome in 17/18 and 19/20 cases,
respectively. Cluster 1 mutants have reduced pol activity but will compete for DNA binding and
cause enzyme stalling when bound. Cluster 2 mutations show reduced DNA binding affinity and
would be unable to compete for DNA binding with a cluster 1 mutant. We predict this would
result in more stalled enzyme complexes than for bound cluster 2 complexes alone. Because
cluster 2 mutants show only modest decreases in pol activity, mtDNA depletion will progress
gradually until a threshold level is reached to give rise to symptoms of POLG syndrome.
The two most common cluster 2 mutations are A467T and P587L, and the latter is always
reported in cis with T251I in patients. The occurrence of two mutations together suggests either a
genetic founder effect (as in Hakonen et al. (46)), or synergic or compensatory effect of the
variants. T251I maps to the Exo domain, but we consider it to be a SNP for the following reasons:
T251 is not a conserved amino acid, is located on the protein surface, and is not in a reasonable
proximity to any functional region. In contrast, P587L is conserved in mammals and fly, is
positioned in the putative DNA binding channel, and two pathogenic mutations, G588D and
124

P589L, are located directly adjacent to P587 on the same hairpin. By considering T251I as a SNP
and excluding it from the functional analysis, P587L can be compared directly to A467T for its
relative pathogenicity. Although no biochemical data is available for P587L, we suggest it will
reduce DNA binding affinity to a much lesser extent than A467T. Direct comparison of the two
mutations is possible in two different cluster 1 mutations. One patient was reported as compound
heterozygous for G848S/A467T with infantile onset (43), whereas another was reported as
compound heterozygous for G848S/P587L with adult onset (26).

Homozygosity for

P587L+T251I results in a late-onset myopathy phenotype ((7) and Suomalainen, unpublished).
The same pattern is observed with K1191N (7, 26).

Severe combination: cluster 1+5
Compound heterozygotes with cluster 1 and cluster 5 mutations were reported showing
that 23/42 mutation combinations had infantile onset POLG syndrome (Figure 35, middle panel).
Such a high occurrence of the most severe form of POLG syndrome, together with the relatively
large number of combinations documented, indicates that the combined defects from cluster 1 and
cluster 5 are highly prone to cause mtDNA depletion at a young age. In particular, the most
common cluster 5 mutation, W748S, manifested infantile onset POLG syndrome in 28/31 patients
when found in combination with cluster 1 mutations. Two particularly severe combinations are
the subcluster combinations 1D + 5B and 1E + 5B; both result in infantile onset POLG syndrome
in >80% of affected patients (Figure 35, lower panel).

Other combinations
Cluster 4 mutations were reported in a total of 17 combinations with cluster 1 and cluster
2 (Figure 35, lower panel). The incidence of infantile onset POLG syndrome was high (15/17),
125

and may indicate that biochemical defects of cluster 4 mutations are uniquely severe in vivo.
Infantile onset was less common for compound heterozygotes with cluster 3 mutations in
combination with cluster 1 (9/17), cluster 2 (10/21) and cluster 5 (3/10) (Figure 36). These results
suggest that cluster 4 mutations are more pathogenic than cluster 3 mutations.

Expected incidence of specific symptoms caused by POLG syndrome can be predicted by
the age of onset.
The clinical manifestations reported for each compound heterozygous patient were
compiled in effort to study the incidence and timeline of symptoms caused by POLG syndrome.
To analyze the clinical descriptions in terms of individual symptoms, it was necessary to group
similar and synonymous symptoms into symptom groups (Table 6). Very common symptoms,
such as ataxia, seizures, hypotonia, and migraines, were left as individual entries because their
sample size was sufficiently large, and these symptoms have been established as hallmarks for
specific POLG syndromes. For other symptom groups, it was necessary to combine different
symptoms into a group of related symptoms based on the affected tissue type resulting in the
observed symptom. For example, the CPEO group includes all synonyms for ophthalmoplegia as
well as diplopia and ptosis, two symptoms that both indicate impaired function of the extraocular
muscles. Neurological disorders are divided into two subcategories; the neuropathy group includes
symptoms caused by damage to peripheral neurons, and the CNS group includes symptoms caused
by damage to neurons of the central nervous system. The developmental delay group includes
symptoms related to an encephalopathy that involve an altered mental state. The myopathy,
hepatopathy, and GI groups involve symptoms that affect or are an indicator of dysfunction in
muscle tissue, the liver, or the gastrointestinal tract, respectively. Figure 37 shows the incidence
of each symptom group in patients as a function of age of onset. The more severe symptoms, such

126

as hepatopathy, hypotonia, and developmental delay have a high incidence among patients with
infantile onset POLG syndrome, whereas less severe symptoms such as CPEO manifest mainly in
patients with adult onset POLG syndrome. These data demonstrate that the symptoms caused by
POLG syndromes represent a continuum of manifestations that become less severe as the age of
onset increases. These syndromes have been characterized in the past according to specific groups
of co-occurring symptoms that present in patients (Figure 38). For example, Alpers syndrome is
characterized by seizures, developmental delay, and liver failure. Figure 38 displays the timeline
of POLG syndromes, and outlines the co-occurring symptoms that define each of the specific
syndromes that have been used to characterize patients with pathogenic POLG mutations.

Figure 37. Age of onset correlation for reported symptoms. Symptoms associated with
POLG syndromes manifest typically at different ages. The symptoms reported in the patients
have been grouped together based on affected tissue types, and ordered according to their ages of
onset. The more severe symptoms and forms of POLG syndromes are at the left end of the
figure, while the less severe symptoms are at the right end. Detailed information of the symptom
grouping is shown in Table 6.

127

Figure 38. Symptoms associated with different POLG syndromes. Depending on the specific
mutations and cluster combinations of the POLG1 gene mutations in individual patients, POLG
syndromes can become symptomatic at different ages and manifest as pathogenic conditions in
different tissue types. The figure shows a continuing spectrum of decreasing symptom severity
from top to bottom, as well as a delayed age of onset from left to right. MCHS, Childhood
myocerebrohepatopathy spectrum; MEMSA, Myoclonic epilepsy myopathy sensory ataxia;
ANS, Autonomic Nervous System Dysfunction; arPEO/ adPEO, Autosomal Recessive/
Dominant Progressive External Ophthalmoplegia.

128

Table 6. Grouping of POLG syndrome symptoms
Symptom
group
Hypotonia
Alpers

Symptoms included

Hypotonic
Alpers syndrome, Assumed Alpers
Altered mental status, Cognitive delay, Dementia, Developmental
Developmental delay, Encephalopathy, Failure to thrive (FTT), Growth
retardation, Lowered consciousness, Psychomotor delay,
delay
Psychomotor regression, Retardation
Cholestasis, Hepatocerebral, Hepatomegaly, Jaundice, Lactic
Hepatopathy acidaemia, Lactic acidosis, Liver cirrhosis, Liver dysfunction,
Liver failure
Cyclic vomiting, Delayed gastric emptying, Diarrhea, GI
GI
dysmotility, GI problems, GI reflux, Pancreatitis, Vomiting
Abrupt onset of seizure, Epilepsia partialis, Epilepsy, Focal
seizures, Hemiparesis, Intractable seizure, Myoclonic seizures,
Seizures
Status epilepticus
Headache/ migraine
Migraines
Dysarthria, Dysphagia, Dystonia, Encephalomyopathy, Metabolic
stroke, Nystagmus, Paralysis, Parkinson's disease, Spasticity,
CNS
Stroke, Tremor
Abnormal muscle histology, Abnormal muscle ultrastructure,
Cox-deficient muscle, Cox-negative, Exercise intolerance,
Myopathy
Mitochondrial myopathy, Muscle weakness, Ragged red fibers
Axonal sensorimotor polyneuropathy, Delayed myelination,
Demyelinating neuropathy, Optic atrophy, Peripheral neuropathy,
Neuropathy
Polyneuropathy, Sensomotor neuropathy
Cerebellar ataxia, Cerebellar atrophy, Movement disorder (ataxia),
Ataxia
Sensory ataxia
Diplopia, External ophthalmoplegia, Ocular myopathy,
CPEO
Ophthalmoplegia, PEO, Ptosis
Areflexia, Arrhythmia, Atrial hypertrophy, Bilateral deafness,
Bilateral lesions of thalami, CPK abnormalities, Cardiopathy,
Chronic bronchitis, Cortical blindness, Distal muscle wasting,
Hearing loss, Hypoglycemia, Hypothyroidism, Ischaemic
Other
episodes, Ketosis, Leigh syndrome, Microcephaly, Occipital
strokes, Ocular bulbar weakness, Proximal weakness, Renal
tubulopathy, Respiratory deficiency, Septicemia, Stroke-like
episodes

129

Patient
number
47
51
148

98
37
162
45
114
86
80
30
168

74

DISCUSSION
Due to the mutational diversity in the POLG1 gene and the variable clinical features of
POLG syndromes, establishment of genotype-phenotype correlations has remained elusive (49).
Furthermore, a number of de novo variants emerge in routine diagnosis laboratories, and
interpretation of their pathogenic role is challenging. We show here that detailed protein structural
and functional studies, with the aid of a crystal structure, combined with knowledge from
experimental and disease mutations can be combined to predict clinical consequences of an
identified variant, and to distinguish likely pathogenic mutations from polymorphic variants. In
this report, we demonstrate clear genotype-phenotype correlations for predicting the severity and
hence age of onset of compound heterozygous POLG syndromes. We refine our clustering model
of Pol γ (1), proposing five functional clusters, each of which comprise subclusters that define
structural regions of the catalytic core of Pol γ. Within these clusters, we can predict the likely
biochemical defects and intrinsic pathogenicity of novel point mutations.

In applying this

clustering protocol to nearly 140 combinations of POLG compound heterozygous mutations, we
showed that each cluster combination can be used to give information of the potential severity of
disease outcome. Strikingly, we were able to predict correctly the age of onset in 90% of infantileonset POLG syndromes via the presence of two mutations from different clusters in compound
heterozygotes of POLG. Furthermore, we show that age of onset can be used to predict the
symptoms that a patient will manifest, and provide a timeline of the disease progression that may
be expected.
Individuals carrying mutations in POLG1 can manifest a disease at any age, depending on
the mutation combination as well as the severity of the Pol γ functional defects (5, 26, 50).
Presumably, the most severe Pol γ defects would be embryonic lethal, and this is likely the reason

130

for example, that nucleotide changes that yield amino acid substitutions in the Mg2+ binding
residues D890 and D1135 have never been reported. We found a similar trend in our clustering
analysis showing a lower occurrence of compound heterozygotes from the same cluster for clusters
1, 3 and 4. The low occurrence of same-cluster compound heterozygotes may indicate that the
majority of potential combinations of this type are lethal, and the lack of complementation supports
our assignment of a common biochemical defect. This does not, however, apply to IP-domain
mutations, which as homozygous are a common cause of severe progressive disorder that has a
later disease-onset in juvenile age or early adulthood. Therefore, combinations of mutations that
cause the most severe form of the POLG syndrome allow embryonic development, but lead to
early infantile death.
We believe that our clustering model of POLG1 mutations can be used as a tool to aid in
the diagnosis of patients with mitochondrial disorders. The POLG1 gene shows considerable
variation, and novel polymorphic changes are common. A valuable existing tool, the polymerase
gamma database (http://tools.niehs.nih.gov/polg/), lists reported mutations, but their pathogenic
role sometimes remains unclear, especially for rare changes. Our prediction tool adds considerable
power via structural analysis for interpretation of the consequences of identified rare variants, and
can provide information for novel changes. Furthermore, it predicts the combinatorial effect of
compound heterozygous changes for a patient, and provides information of typical clinical
manifestations and ages of onset. We note that these predictions may not always be accurate, both
because of the variability in disease manifestation introduced by genetic background, and the
possibility that a number of the patients reported with POLG syndrome may also have other
inherited mutations in their mitochondrial or nuclear DNAs in addition to the reported mutations
in the POLG gene; although we excluded known cases of digenic mitochondrial diseases in this
131

study, the latter possibility remains problematic and would be addressed by whole exome
sequencing. Despite these potential drawbacks, we believe that our clustering protocol provides
additional information that will prove important not only for diagnosis and genetic counseling, but
also for treatment: e.g., the epilepsy drug valproate is toxic to the liver of patients with POLG1
mutations (51), and as epilepsy is a common initial symptom in children with POLG diseases,
many patients have died before receiving a liver transplant. The ability to predict pathogenic
likelihood for new POLG variants will be important when considering valproate as a treatment in
severe epilepsies. Finally, because the predictions are based on reported patient data, the everincreasing number of documented POLG1 mutations will serve to enhance the diagnostic power
of the clustering protocol.

132

APPENDIX

133

APPENDIX

Subcluster Definitions
Cluster 1
Location in structure: Pol active site and environs
Biochemical phenotype: Reduced pol activity caused by a defect in the pol rate
Description: Cluster 1 mutations are predicted to cause a primary defect in pol activity, and affect
residues involved directly in catalysis, or indirectly by affecting architectural residues that may
disrupt the position of catalytic residues. Catalytic residues include those binding the two Mg2+
ions, those that make contact with the incoming dNTP, and those that make contact with the first,
second and third nucleotide pairs in the nascent dsDNA, which are critical for correct positioning
of the substrate in the pol active site (Euro et al, 2011; Farnum et al, this report).

Subcluster 1A
Location in structure: Outer layer of the holoenzyme, contacts subcluster 1G and is adjacent to
subcluster 1F
Biochemical phenotype: Mild defect in pol activity or diminished stability/ solubility of the
polypeptide
Description: No biochemical data are available for this region. We predict an architectural role
via stabilization of subcluster 1G.
Recessive/ dominant mutation predictions: Recessive

134

Subcluster 1B
Location in structure: Outer layer of the holoenzyme, contacts subclusters 1C, 1G, and 1F
Biochemical phenotype: Mild defect in pol activity or diminished stability/ solubility of the
polypeptide
Description: No biochemical data are available for this region. We predict an architectural role
via stabilization of subclusters 1C, 1G, and 1F.
Recessive/ dominant mutation predictions: Recessive

Subcluster 1C
Location in structure: Mostly solvent exposed, contacts subclusters 1B, 1D, 1F, and 1G;
equivalent to the G-helix of other family A polymerases
Biochemical phenotype: Mild defect in pol activity or diminished stability/ solubility of the
polypeptide
Description: No biochemical data are available for this region. This subcluster is adjacent to motif
2 and motif 6 in the pol domain, and we propose it serves an architectural role that contributes to
pol activity indirectly, by stabilizing subclusters 1B, 1D, 1F, and 1G. Patient data shows these
mutations to be recessive.
Recessive/ dominant mutation predictions: Recessive

Subcluster 1D
Location in structure: This subcluster forms a large portion of the pol active site and contains
two highly conserved motifs that are found in all family A polymerases: the RR loop (motif 2) and
motif A (Loh and Loeb, 2005).

135

Biochemical phenotype: Reduced pol rate, possibly reduced DNA binding affinity
Description: Motif 2 and the Pol A motif are both critical for catalysis but contribute to catalysis
by different mechanisms. Therefore, a mutation in subcluster 1D should be evaluated by the motif
to which it is closest in primary sequence.
Motif 2 (residues 845-863) In Pol γ, motif 2 in the pol domain has been termed the "RR loop"
because of two sequential arginines (R852 and R853) located on the tip of the loop, whereas family
A polymerases have only a single arginine (equivalent to R853) (Euro et al, 2011). Overall, motif
2 is critical for binding correctly base-paired primer-template DNA in the minor groove, and to
stabilize the template DNA backbone in the active site (Loh and Loeb, 2005). These residues form
specific contacts with DNA in the pol active site and are major determinants for the catalytic
efficiency and fidelity of DNA synthesis by Pol γ. Motif 2 mutations should always cause reduced
pol activity but may also cause a DNA binding defect depending on the residue, such as the 5-fold
defects reported for G848S and R852C (Kasiviswanathan et al, 2009). The DNA binding affinity
of T851A, R853Q, Q879H, T885S may be decreased slightly, but the reduction in pol activity is
much greater (Kasiviswanathan et al, 2009). Overall, it is unlikely for motif 2 mutations to present
as dominant mutations because these mutant forms of Pol γ would not compete effectively with
wild type Pol γ for DNA binding.
Pol A motif (residues 887-896) The Pol A motif of family A DNA polymerases forms the site of
catalysis where it contacts closely the primer strand and binds the Mg2+ ion required for the
chemical steps of nucleotide polymerization (D890 in Pol γ). Additionally, a residue of the Pol A
motif discriminates against ribonucleotide incorporation (E895 in Pol γ), and mutation causes
severe pol defects in addition to increased ribonucleotide incorporation (Astatke et al, 1998).
Mutations in the Pol A motif at either D890 or E895 in Pol γ would likely manifest a dominant
136

lethal phenotype, and the only patient carrying the E895G mutation to date died immediately after
birth (Spinazzola et al, 2009). In contrast to these two critical residues, adjacent residues within
the Pol A motif play lesser roles in pol activity and mutations in them are observed more
frequently, resulting in reduced pol activity without a significant effect on DNA binding affinity,
and thus they can produce a dominant phenotype. Overall, though dominant mutations are fairly
rare for subcluster 1D, novel mutations affecting residues of the Pol A motif would present a
moderate risk for dominant inheritance.
Recessive/ dominant mutation predictions: Moderate risk of dominant mutations within the Pol
A motif (residues 887-896)

Subcluster 1E
Location in structure: This subcluster comprises most of the fingers subdomain of the pol
domain, including the O-helix and the Pol B motif (Loh and Loeb, 2005).
Biochemical phenotype: Reduced pol rate, and altered dNTP binding affinity and fidelity
Pol B motif (residues 943-958) The conserved amino acid residues on the O-helix are known as
the Pol B motif, and function to establish specific contacts with correctly base paired dNTP in the
closed conformation. POLG mutations that disrupt the specific contacts with the incoming dNTP
(H932Y, R943H, K947R, Y951N and Y955C) will reduce fidelity and increase K m (dNTP),
without affecting DNA binding affinity (Graziewicz et al, 2004; Batabyal et al, 2010). Mutations
affecting this site will most likely be dominant because they are capable of competing for dNTP
binding with wild type Pol γ, but are unable to polymerize nucleotides effectively, and are
predicted to cause mtDNA damage that is associated with enzyme stalling (Atanassova et al,
2011).
137

Recessive/ dominant mutation predictions: High risk of dominant mutations in the Pol B motif
(residues 943-958); subcluster 1E residues outside this motif can be dominant, but the risk is much
lower.

Subcluster 1F
Location in structure: This subcluster forms a segment of the pol active site and contains two
highly conserved motifs that are found in all family A polymerases, the Pol C motif and motif 6
(Loh and Loeb, 2005).
Biochemical phenotype: Reduced pol rate, possibly reduced DNA binding affinity
Description: Motif 6 (residues 1097-1110) is located on the Q-helix and binds correctly base
paired template strand and the minor groove of the primer-template DNA. The Pol C motif
(residues 1134-1141) binds Mg2+ via residue D1135, whereas H1134 and E1136 contact the
primer strand. Mutations in the Pol C motif will affect the rate of the chemical step of dNTP
addition to the primer terminus.
Recessive/ dominant mutation predictions: Moderate risk of dominant mutations in the Pol C
motif (residues 1134-1141)

Subcluster 1G
Location in structure: Subcluster 1G (residues 1157-1196) maps to a C-terminal region of the
palm subdomain, and forms an anti-parallel beta strand adjacent to the Pol A and Pol C motifs
(Loh and Loeb, 2005).
Biochemical phenotype: Mild defect in pol activity or diminished stability/ solubility of the
polypeptide.
138

Description: This subcluster is adjacent to motifs A and C, and we propose it serves an
architectural role that contributes to pol activity indirectly, by stabilizing subclusters 1D and 1F.
Only K1191 is predicted to contact the primer terminus, whereas the remainder of the subcluster
1G residues serve an architectural role. Patient data shows these mutations to be recessive.
Recessive/ dominant mutation predictions: Recessive

Cluster 2
Location in structure: Upstream DNA binding channel of the spacer domain
Biochemical phenotype: Reduced DNA binding affinity
Affected step in polymerization pathway: The kon and koff rates will be altered by cluster 2
mutations, which will manifest as an increase in the Kd of primer-template DNA binding.
Description: Cluster 2 mutations are predicted to cause a primary defect in the DNA binding
affinity of Pol γ to the primer-template DNA to form the binary complex. They affect residues of
the intrinsic processivity subdomain (IP) and accessory interacting domain (AID) of the spacer
domain that form the upstream DNA binding channel, and which enhance DNA binding affinity
via contacts with nucleotides upstream of the third base pair of primer template DNA.
Recessive/ dominant mutation predictions: Recessive

Subcluster 2A
Location in structure: Subcluster 2A maps to a region of the thumb subdomain of the pol domain
at the accessory subunit interface where A467T, N468D and L463F are positioned.
Biochemical phenotype: Reduced DNA binding affinity

139

Description: Extensive biochemical analysis of the spacer domain in general has demonstrated
that the hydrophobic core of the IP subdomain is critical for shaping the DNA binding channel
wall, and mutations are predicted to perturb the channel to block DNA from entering, which results
in reduced DNA binding affinity.
Recessive/ dominant mutation predictions: Recessive

Subcluster 2B
Location in structure: Subcluster 2B maps to the region of the AID that is predicted to contact
the upstream DNA duplex.
Biochemical phenotype: Reduced DNA binding affinity
Description: The AID is stabilized in the presence of the accessory subunit and enhances DNA
binding affinity through many positively charged residues. Subcluster 2B maps to the region of
the AID containing the positively charged residues, and mutations may alter their interaction with
bound DNA.
Recessive/ dominant mutation predictions: Recessive

Subcluster 2C
Location in structure: Subcluster 2C contains motif 1, which in Pol I was shown to fold into a
loop that binds DNA in the channel (Loh and Loeb, 2005). Motif 1, together with subcluster 2D,
form the major face of the putative DNA binding channel.
Biochemical phenotype: Reduced DNA binding affinity
Description: Extensive biochemical analysis of the spacer domain in general has demonstrated
that the hydrophobic core of the IP subdomain is critical for shaping the DNA binding channel
140

wall, and mutations are predicted to perturb the channel to block DNA from entering, which is
then observed as reduced DNA binding affinity.
Recessive/ dominant mutation predictions: Recessive

Subcluster 2D
Location in structure: Subcluster 2D, together with subcluster 2C, form the major face of the
putative DNA binding channel.
Biochemical phenotype: Reduced DNA binding affinity
Description: Extensive biochemical analysis of the spacer domain in general has demonstrated
that the hydrophobic core of the IP subdomain is critical for shaping the DNA binding channel
wall, and mutations are predicted to perturb the channel to block DNA from entering, which is
then observed as reduced DNA binding affinity.
Recessive/ dominant mutation predictions: Recessive

Cluster 3
Location in structure: Cluster 3 defines a group of structural modules that are located between
the pol and exo active sites: a region near the exo active site (3A), the orienter module (3B), a
segment of the thumb subdomain (3C), and the partitioning loop (3D).
Biochemical phenotype: Predicted to affect the pol: exo activity ratio and may have variable
defects in DNA binding affinity
Description: Cluster 3 mutations are predicted to affect the pol: exo activity ratio; they have been
reported to cause DNA binding defects in most biochemical variants studied.
Recessive/ dominant mutation predictions: Recessive

141

Subcluster 3A
Location in structure: A region near the exo active site that comprises the Exo II motif
Biochemical phenotype: Reduced exo activity, wild type pol activity
Description: The Exo II motif has been demonstrated to alter the pol: exo activity ratio in
biochemical variants of Pol I by decreasing the affinity of the primer strand for the exo active site,
consistent with the decrease in exo activity and wild type pol activity observed for the R233W
variant in yeast Pol γ (H277 in human Pol γ) (Foury and Szczepanowska, 2011).
Recessive/ dominant mutation predictions: Recessive

Subcluster 3B
Location in structure: A helix-coil-helix module (residues 295-312) located in the Exo domain
that has been termed the "orienter" module
Biochemical phenotype: Reduced DNA binding affinity, reduced pol activity, possible increase
in exo activity
Description: All biochemical variants of the orienter module have shown reduced DNA-binding
affinity and reduced pol activity and in addition, the L304R variant showed a significant increase
in exo activity (Foury and Szczepanowska, 2010).
Recessive/ dominant mutation predictions: Recessive

Subcluster 3C
Location in structure: A segment of the thumb subdomain of the pol domain located on the inner
face of the DNA binding channel between the pol and exo active sites

142

Biochemical phenotype: Increased exo activity, decreased pol activity, possible decrease in DNA
binding affinity
Description: Mutations in subcluster 3C have been reported to cause DNA binding defects in most
biochemical variants studied. Notably, an SYW triple alanine substitution in recombinant fly Pol
γ, which was found to exhibit increased exo activity with both decreased pol activity and DNA
binding affinity (Luo and Kaguni, 2005), maps adjacent to subcluster 3C in human Pol γ (residues
799

SFW801). Biochemical study of a variant of yeast Pol γ M602I (R802 in Hs Pol γ) showed

wild type DNA binding affinity, decreased pol activity, and increased exo activity (Foury and
Szczepanowska, 2011).
Recessive/ dominant mutation predictions: Recessive

Subcluster 3D
Location in structure: The partitioning loop, which is a novel structural module conserved in
Pol γ (residues 1050-1095) that is located between the fingers and palm subdomains of the pol
domain, and is not present in any other known DNA polymerase (Euro et al, 2011)
Biochemical phenotype: Predicted to alter the pol: exo ratio and possibly DNA binding affinity
Description: To date, no biochemical data are available for residues of the partitioning loop,
although it has been shown that yeast strains homozygous for a mutation equivalent to G1051R in
human Pol γ cause a 10-fold increase in point mutational frequency in vivo (Baruffini et al, 2007).
Recessive/ dominant mutation predictions: Recessive

143

Cluster 4
Location in structure: These mutations map to the exo domain along the distal accessory subunit
interface.
Biochemical phenotype: Increase in exo activity, decrease in pol activity
Description: Biochemical studies of a human R232G variant in reconstituted holoenzyme form
showed a decrease in pol rate and an increase in exo activity, with unchanged DNA binding affinity
that derives from loss of a direct stimulation of pol activity by the distal accessory subunit (Lee et
al, 2010).

We propose that the other residues of cluster 4 will have similar biochemical

characteristics.
Recessive/ dominant mutation predictions: Recessive

Cluster 5
Location in structure: Located in the periphery of the IP subdomain of the spacer domain, distant
from the DNA binding channel
Biochemical phenotype: None observed
Description: Cluster 5 mutations are located in the periphery of the IP subdomain and have not
been shown to cause a biochemical defect in Pol γ per se. We predict that these mutations may
affect as yet undetermined protein-protein interactions in vivo (Euro et al, 2011).
Recessive/ dominant mutation predictions: Recessive

144

BIBLIOGRAPHY

145

BIBLIOGRAPHY

1.
Euro, L., Farnum, G.A., Palin, E., Suomalainen, A. and Kaguni, L.S. (2011) Clustering of
Alpers disease mutations and catalytic defects in biochemical variants reveal new features of
molecular mechanism of the human mitochondrial replicase, Pol {gamma}. Nucleic Acids Res.
2.
Ylikallio, E. and Suomalainen, A. (2011) Mechanisms of mitochondrial diseases. Ann
Med.
3.
Kaguni, L.S. (2004) DNA polymerase gamma, the mitochondrial replicase. Annu Rev
Biochem, 73, 293-320.
4.
Yakubovskaya, E., Chen, Z., Carrodeguas, J.A., Kisker, C. and Bogenhagen, D.F. (2006)
Functional human mitochondrial DNA polymerase gamma forms a heterotrimer. J Biol Chem,
281, 374-82.
5.
Saneto, R.P. and Naviaux, R.K. (2010) Polymerase gamma disease through the ages. Dev
Disabil Res Rev, 16, 163-74.
6.
Nguyen, K.V., Sharief, F.S., Chan, S.S., Copeland, W.C. and Naviaux, R.K. (2006)
Molecular diagnosis of Alpers syndrome. J Hepatol, 45, 108-16.
7.
Horvath, R., Hudson, G., Ferrari, G., Futterer, N., Ahola, S., Lamantea, E., Prokisch, H.,
Lochmuller, H., McFarland, R., Ramesh, V. et al. (2006) Phenotypic spectrum associated with
mutations of the mitochondrial polymerase gamma gene. Brain, 129, 1674-84.
8.
Lee, Y.S., Kennedy, W.D. and Yin, Y.W. (2009) Structural insight into processive human
mitochondrial DNA synthesis and disease-related polymerase mutations. Cell, 139, 312-24.
9.
Lee, Y.S., Lee, S., Demeler, B., Molineux, I.J., Johnson, K.A. and Yin, Y.W. (2009)
Each monomer of the dimeric accessory protein for human mitochondrial DNA polymerase has a
distinct role in conferring processivity. J Biol Chem, 285, 1490-9.
10.
Singh, K. and Modak, M.J. (2005) Contribution of polar residues of the J-helix in the 3'5' exonuclease activity of Escherichia coli DNA polymerase I (Klenow fragment): Q677
regulates the removal of terminal mismatch. Biochemistry, 44, 8101-10.
11.
McCain, M.D., Meyer, A.S., Schultz, S.S., Glekas, A. and Spratt, T.E. (2005) Fidelity of
mispair formation and mispair extension is dependent on the interaction between the minor
groove of the primer terminus and Arg668 of DNA polymerase I of Escherichia coli.
Biochemistry, 44, 5647-59.
12.
Loh, E. and Loeb, L.A. (2005) Mutability of DNA polymerase I: implications for the
creation of mutant DNA polymerases. DNA Repair (Amst), 4, 1390-8.
13.
Singh, K. and Modak, M.J. (2003) Presence of 18-A long hydrogen bond track in the
active site of Escherichia coli DNA polymerase I (Klenow fragment). Its requirement in the
stabilization of enzyme-template-primer complex. J Biol Chem, 278, 11289-302.

146

14.
Patel, P.H., Suzuki, M., Adman, E., Shinkai, A. and Loeb, L.A. (2001) Prokaryotic DNA
polymerase I: evolution, structure, and "base flipping" mechanism for nucleotide selection. J Mol
Biol, 308, 823-37.
15.
Doublie, S., Tabor, S., Long, A.M., Richardson, C.C. and Ellenberger, T. (1998) Crystal
structure of a bacteriophage T7 DNA replication complex at 2.2 A resolution. Nature, 391, 2518.
16.
Johnson, A.A., Tsai, Y., Graves, S.W. and Johnson, K.A. (2000) Human mitochondrial
DNA polymerase holoenzyme: reconstitution and characterization. Biochemistry, 39, 1702-8.
17.
Luo, N. and Kaguni, L.S. (2005) Mutations in the spacer region of Drosophila
mitochondrial DNA polymerase affect DNA binding, processivity, and the balance between Pol
and Exo function. J Biol Chem, 280, 2491-7.
18.
Szczepanowska, K. and Foury, F. (2010) A cluster of pathogenic mutations in the 3'-5'
exonuclease domain of DNA polymerase gamma defines a novel module coupling DNA
synthesis and degradation. Hum Mol Genet, 19, 3516-29.
19.
Graziewicz, M.A., Longley, M.J., Bienstock, R.J., Zeviani, M. and Copeland, W.C.
(2004) Structure-function defects of human mitochondrial DNA polymerase in autosomal
dominant progressive external ophthalmoplegia. Nat Struct Mol Biol, 11, 770-6.
20.
Baruffini, E., Ferrero, I. and Foury, F. (2007) Mitochondrial DNA defects in
Saccharomyces cerevisiae caused by functional interactions between DNA polymerase gamma
mutations associated with disease in human. Biochim Biophys Acta, 1772, 1225-35.
21.
Lee, Y.S., Johnson, K.A., Molineux, I.J. and Yin, Y.W. (2010) A single mutation in
human mitochondrial DNA polymerase Pol gammaA affects both polymerization and
proofreading activities of only the holoenzyme. J Biol Chem, 285, 28105-16.
22.
Luoma, P.T., Luo, N., Loscher, W.N., Farr, C.L., Horvath, R., Wanschitz, J., Kiechl, S.,
Kaguni, L.S. and Suomalainen, A. (2005) Functional defects due to spacer-region mutations of
human mitochondrial DNA polymerase in a family with an ataxia-myopathy syndrome. Hum
Mol Genet, 14, 1907-20.
23.
Ferreira, M., Evangelista, T., Almeida, L.S., Martins, J., Macario, M.C., Martins, E.,
Moleirinho, A., Azevedo, L., Vilarinho, L. and Santorelli, F.M. (2011) Relative frequency of
known causes of multiple mtDNA deletions: two novel POLG mutations. Neuromuscul Disord,
21, 483-8.
24.
Sato, K., Yabe, I., Yaguchi, H., Nakano, F., Kunieda, Y., Saitoh, S. and Sasaki, H. (2011)
Genetic analysis of two Japanese families with progressive external ophthalmoplegia and
parkinsonism. J Neurol, 258, 1327-32.
25.
Tang, S., Dimberg, E.L., Milone, M. and Wong, L.J. (2011) Mitochondrial
neurogastrointestinal encephalomyopathy (MNGIE)-like phenotype: an expanded clinical
spectrum of POLG1 mutations. J Neurol.
26.
Tang, S., Wang, J., Lee, N.C., Milone, M., Halberg, M.C., Schmitt, E.S., Craigen, W.J.,
Zhang, W. and Wong, L.J. (2011) Mitochondrial DNA polymerase gamma mutations: an ever
expanding molecular and clinical spectrum. J Med Genet, 48, 669-81.

147

27.
Brieba, L.G., Eichman, B.F., Kokoska, R.J., Doublie, S., Kunkel, T.A. and Ellenberger,
T. (2004) Structural basis for the dual coding potential of 8-oxoguanosine by a high-fidelity
DNA polymerase. EMBO J, 23, 3452-61.
28.
Baruffini, E., Horvath, R., Dallabona, C., Czermin, B., Lamantea, E., Bindoff, L.,
Invernizzi, F., Ferrero, I., Zeviani, M. and Lodi, T. (2010) Predicting the contribution of novel
POLG mutations to human disease through analysis in yeast model. Mitochondrion, 11, 182-90.
29.
Khan, A., Trevenen, C., Wei, X.C., Sarnat, H.B., Payne, E. and Kirton, A. (2011) Alpers
Syndrome: The Natural History of a Case Highlighting Neuroimaging, Neuropathology, and Fat
Metabolism. J Child Neurol.
30.
Li, Y., Korolev, S. and Waksman, G. (1998) Crystal structures of open and closed forms
of binary and ternary complexes of the large fragment of Thermus aquaticus DNA polymerase I:
structural basis for nucleotide incorporation. EMBO J, 17, 7514-25.
31.
Batabyal, D., McKenzie, J.L. and Johnson, K.A. (2010) Role of histidine 932 of the
human mitochondrial DNA polymerase in nucleotide discrimination and inherited disease. J Biol
Chem, 285, 34191-201.
32.
Atanassova, N., Fuste, J.M., Wanrooij, S., Macao, B., Goffart, S., Backstrom, S., Farge,
G., Khvorostov, I., Larsson, N.G., Spelbrink, J.N. et al. (2011) Sequence-specific stalling of
DNA polymerase gamma and the effects of mutations causing progressive ophthalmoplegia.
Hum Mol Genet, 20, 1212-23.
33.
Astatke, M., Ng, K., Grindley, N.D. and Joyce, C.M. (1998) A single side chain prevents
Escherichia coli DNA polymerase I (Klenow fragment) from incorporating ribonucleotides. Proc
Natl Acad Sci U S A, 95, 3402-7.
34.
Spinazzola, A., Invernizzi, F., Carrara, F., Lamantea, E., Donati, A., Dirocco, M.,
Giordano, I., Meznaric-Petrusa, M., Baruffini, E., Ferrero, I. et al. (2009) Clinical and molecular
features of mitochondrial DNA depletion syndromes. J Inherit Metab Dis, 32, 143-58.
35.
Kasiviswanathan, R., Longley, M.J., Chan, S.S. and Copeland, W.C. (2009) Disease
mutations in the human mitochondrial DNA polymerase thumb subdomain impart severe defects
in mitochondrial DNA replication. J Biol Chem, 284, 19501-10.
36.
Li, Y., Kong, Y., Korolev, S. and Waksman, G. (1998) Crystal structures of the Klenow
fragment of Thermus aquaticus DNA polymerase I complexed with deoxyribonucleoside
triphosphates. Protein Sci, 7, 1116-23.
37.
Foury, F. and Szczepanowska, K. (2011) Antimutator Alleles of Yeast DNA Polymerase
Gamma Modulate the Balance between DNA Synthesis and Excision. PLoS One, 6, e27847.
38.
Suomalainen, A. and Isohanni, P. (2010) Mitochondrial DNA depletion syndromes-many genes, common mechanisms. Neuromuscul Disord, 20, 429-37.
39.
Srivastava, S. and Moraes, C.T. (2005) Double-strand breaks of mouse muscle mtDNA
promote large deletions similar to multiple mtDNA deletions in humans. Hum Mol Genet, 14,
893-902.
40.
Fukui, H. and Moraes, C.T. (2009) Mechanisms of formation and accumulation of
mitochondrial DNA deletions in aging neurons. Hum Mol Genet, 18, 1028-36.
148

41.
Wanrooij, S., Goffart, S., Pohjoismaki, J.L., Yasukawa, T. and Spelbrink, J.N. (2007)
Expression of catalytic mutants of the mtDNA helicase Twinkle and polymerase POLG causes
distinct replication stalling phenotypes. Nucleic Acids Res, 35, 3238-51.
42.
Wong, L.J., Naviaux, R.K., Brunetti-Pierri, N., Zhang, Q., Schmitt, E.S., Truong, C.,
Milone, M., Cohen, B.H., Wical, B., Ganesh, J. et al. (2008) Molecular and clinical genetics of
mitochondrial diseases due to POLG mutations. Hum Mutat, 29, E150-72.
43.
Stewart, J.D., Tennant, S., Powell, H., Pyle, A., Blakely, E.L., He, L., Hudson, G.,
Roberts, M., du Plessis, D., Gow, D. et al. (2009) Novel POLG1 mutations associated with
neuromuscular and liver phenotypes in adults and children. J Med Genet, 46, 209-14.
44.
Palin, E.J., Lesonen, A., Farr, C.L., Euro, L., Suomalainen, A. and Kaguni, L.S. (2010)
Functional analysis of H. sapiens DNA polymerase gamma spacer mutation W748S with and
without common variant E1143G. Biochim Biophys Acta, 1802, 545-51.
45.
Uusimaa, J., Hinttala, R., Rantala, H., Paivarinta, M., Herva, R., Roytta, M., Soini, H.,
Moilanen, J.S., Remes, A.M., Hassinen, I.E. et al. (2008) Homozygous W748S mutation in the
POLG1 gene in patients with juvenile-onset Alpers syndrome and status epilepticus. Epilepsia,
49, 1038-45.
46.
Hakonen, A.H., Heiskanen, S., Juvonen, V., Lappalainen, I., Luoma, P.T., Rantamaki,
M., Goethem, G.V., Lofgren, A., Hackman, P., Paetau, A. et al. (2005) Mitochondrial DNA
polymerase W748S mutation: a common cause of autosomal recessive ataxia with ancient
European origin. Am J Hum Genet, 77, 430-41.
47.
Winterthun, S., Ferrari, G., He, L., Taylor, R.W., Zeviani, M., Turnbull, D.M., Engelsen,
B.A., Moen, G. and Bindoff, L.A. (2005) Autosomal recessive mitochondrial ataxic syndrome
due to mitochondrial polymerase gamma mutations. Neurology, 64, 1204-8.
48.
Van Goethem, G., Luoma, P., Rantamaki, M., Al Memar, A., Kaakkola, S., Hackman, P.,
Krahe, R., Lofgren, A., Martin, J.J., De Jonghe, P. et al. (2004) POLG mutations in
neurodegenerative disorders with ataxia but no muscle involvement. Neurology, 63, 1251-7.
49.
Milone, M., Benarroch, E.E. and Wong, L.J. (2011) POLG-related disorders: Defects of
the nuclear and mitochondrial genome interaction. Neurology, 77, 1847-52.
50.
Blok, M.J., van den Bosch, B.J., Jongen, E., Hendrickx, A., de Die-Smulders, C.E.,
Hoogendijk, J.E., Brusse, E., de Visser, M., Poll-The, B.T., Bierau, J. et al. (2009) The unfolding
clinical spectrum of POLG mutations. J Med Genet, 46, 776-85.
51.
Stewart, J.D., Horvath, R., Baruffini, E., Ferrero, I., Bulst, S., Watkins, P.B., Fontana,
R.J., Day, C.P. and Chinnery, P.F. (2010) Polymerase gamma gene POLG determines the risk of
sodium valproate-induced liver toxicity. Hepatology, 52, 1791-6.

149