Articles in PresS. Am J Physiol Endocrinol Metab (July 5, 2005). doi:10.1152/ajpendo.00600.

2004

Parker et al page 1

A common mitchondrial DNA variant is associated with thinness in mothers and their
20 year old offspring

Ellen Parker, DIW Phillips^, Richard A Cockington @, Carole Cull+, Jo Poulton*

* Corresponding author

Nuffield Department of Obstetrics and Gynaecology, The Women’s Centre, John Radcliffe Hospital,
Oxford OX3 9DU.

^ MRC Environmental Epidemiology Unit, University of Southampton

@ Department of Child and Adolescent Development and Rehabilitation, Women's and Children's
Hospital, Adelaide, South Australia.

+ Diabetes Trials Unit, Oxford Centre for Diabetes, Endocrinology and Metabolism, Churchill
Hospital, Oxford OX3 7LJ

Telephone (44) 1865 221067, fax (44) 1865 769141, Joanna.poulton@obs-gyn.ox.ac.uk

Key words

Thinness

Mitochondrial DNA

16189 variant

Placenta

Birth weight

Running Head: 16189 mtDNA variant and thinness

Copyright © 2005 by the American Physiological Society.

 Parker et al page 2

ABSTRACT

Background

A common mitochondrial DNA variant, that is maternally inherited, the 16189 variant, is associated with

type 2 diabetes and thinness at birth. In order to elucidate the association of the variant with thinness, we

studied the 16189 variant in a well characterised Australian cohort (n=161) who were followed up from

birth to age 20 years.

Methods

PCR analysis and mtDNA haplotyping was carried out on DNA from 161 offspring from consecutive,

normal singleton pregnancies were followed from birth to ages 20.

Results

The 16189 mt DNA variant was present in 14 of the 161 20 year olds (8.7%). Both the mothers with the

16189 variant and their 20 year old offspring were thinner than those without. Median (IQR) BMI was 21.9

(20.4 to 22.9) in mothers with the variant compared with 23.5 (21.4 to 26.6) in those without (p=0.013), and

22.2 (21.1 to 23.8) in 20 year olds with the variant compared with 22.7 (20.8 to 25.6) in those without

(p=0.019). The 16189 variant was also associated with a high placental weight and high placental to birth

weight ratio (p=0.051 and p=0.0024 respectively). Insulin sensitivity was normal in 20 year olds with the

16189 variant. This contrasts with 20 year olds who did not have the variant but who had been thin or small

at birth, who had normal BMI and normal placental to birth weight ratio, but were insulin resistant.

Conclusion

This study suggests that 16189 mt DNA variant is associated with maternally inherited thinness in young

adults. This may be mediated by effects on mtDNA replication and thence placental function. Further

research is required to confirm these hypotheses.

 Parker et al page 3

INTRODUCTION

Obesity and fat distribution are important determinants of type 2 diabetes and early cardiovascular

mortality (18). Mitochondrial function is emerging as an important contributor to the risk of diabetes

(19, 23, 25) and other diseases related to aging (24, 35). Transcription factors regulating the formation

and activity of fat cells (6) are also regulators of mitochondrial function (14). Drugs used for treating

AIDS which result in mtDNA depletion (due to inhibition of the mtDNA DNA polymerase) affect

adipocyte differentiation both in vivo (22) and in tissue culture. Patients with mtDNA disease are often

thin and may have abnormal fat (36). Furthermore, we have shown that a common mitochondrial

polymorphism, that we have termed the 16189 variant, is associated with a low Ponderal index in

babies who are small in utero and subsequently “catch up” (4).

Diabetes can be precipitated by exposure to mitochondrial poisons (37) and by down-regulation of

mitochondrial function in a mouse knock out (32), because ATP is critically important to both insulin

production and release (2, 33). Hence, insulin secretion is impaired in maternally inherited diabetes

and/or deafness (MIDD) (34), caused by the 3243G:C point mutation of mitochondrial DNA (29). In

addition, beta cell numbers may be reduced in mtDNA disorders (27). We previously demonstrated

that a common variant of mtDNA at bp 16189 is associated with type 2 diabetes in a case control study

(26). We also showed that the variant was associated with decreased Ponderal index (weight/(height)2)

and impaired glucose tolerance in adults who had who had gained weight rapidly after birth. This is

biologically credible (7) and suggests that the 16189 variant exacerbates the effect of impaired

intrauterine nutrition on both thinness and on subsequent glucose tolerance (4),

 Parker et al page 4

The aim of this study was to see if the 16189 variant was associated with reduced BMI in mothers and

their offspring from birth to 20 years in a well characterised cohort (8), and to see if this was related to

placental weight or glucose tolerance.

 Parker et al page 5

SUBJECTS, MATERIALS AND METHODS

DNA was available on 161 out of a total of 764 individuals (21%) from the Adelaide Children’s Hospital

Family Heart Survey (8). All 161 were offspring from singleton pregnancies who were followed and

underwent measurment of BMI (21) and an intravenous glucose tolerance test (IVGTT) at age 20 years.

The BMI of the mothers had been measured when their offspring were aged 8 years.

For the purposes of the current study, the lightest one third of the whole population at birth (that is males

weighing <3.32 and females <3.18 kg ) were defined as low birth weight and the higher two tertiles as

normal weight. DNA was analysed using PCR of a 506 bp region of the large non-coding region including

the first hypervariable region (HV1) (26). Restriction digestion of the PCR product was used as previously

(26) to identify individuals with the 16189 variant. All PCR products in which the Mnl1 site loss generated

by the 16189 variant was detected were sequenced in both directions as previously in order to confirm the

presence of the homopolymeric C tract and to determine the mtDNA haplotype (30). As previously, we

defined the 16189 variant as the T16189C transition along with an uninterrupted homopolymeric C tract.

This definition excludes those with both the T16189C transition and homoplasmy for the C16186T or

C16192T transition (Poulton, JMG in press), and could thus be termed the mtDNA 16184-93 PolyC tract

(Chinnery, JMG in press).

Statistical analyses were performed using SAS v 8.2 (SAS Institute; Cary, NC, USA). Continuous data are

reported as mean (standard deviation) if normally distributed, geometric mean [1 standard deviation

interval] if log-normally distributed, or median [interquartile range], and categorical data as percentages.

For normally and log-normally distributed continuous variables, comparison of multiple means was by t-
 Parker et al page 6

tests, and adjusted p-values were obtained from a generalised linear model (GLM). For non-normally

distributed continuous variables were log transformed before analysis. For categorical variables a chi-

squared test was used. P<0.05 was taken as conventionally significant. To allow for multiple testing

P<0.0038 would be required.

RESULTS

Fourteen individuals, out of the 161 individuals had the 16189 variant (prevalence 8.7%) confirmed by

restriction enzyme analysis and sequencing. BMI of mothers of offspring with the 16189 variant were

significantly lower (p=0.013). As expected, there was a highly significant association between

maternal weight and birth weight (rs=0.23, p=0.0051, partial correlation controlling for sex and length

of gestation), between BMI and maternal BMI (rs=0.27, p=0.0006, partial correlation of BMI on

maternal BMI controlling for sex, and the presence of the variant) and between maternal BMI and birth

weight (rs=0.19, p=0.014, partial correlation controlling for sex). Maternal BMI was a major

determinant of BMI at aged 20 and had considerably more impact than did birth weight. The

proportion of variance in BMI at aged 20 accounted for by maternal BMI controlling for sex

(difference in R-squared) was 80%, while the proportion accounted for by birthweight was 2%.

The presence of the 16189 variant was associated with BMI at aged 20, placental weight, placental to

birth weight ratio and length of gestation (p=0.019, 0.051, 0.0024 and 0.033 respectively, Table 1). The

associations of the presence of the variant with placental weight, placental weight to birth weight ratio

and maternal BMI were independent of gestational age, sex, maternal age, BMI at age 20 and birth

weight, while that with gestational age was independent of sex (Table 1). The association with BMI at

aged 20 was not independent of sex or gestational age.

 Parker et al page 7

Plasma insulin was lower in individuals with the 16189 variant than those without, at all time points during

the IVGTT (see figure 1), but because individuals with the 16189 variant were significantly thinner this did

not reach clear statistical significance. There was no clear difference in insulin sensitivity assessed from the

IVGTT data or in beta cell function as assessed with the HOMA2 model (Homeostatic Model Assessment

(17)).

Sequence analysis demonstrated 12 distinct mitochondrial haplotypes in individuals with the 16189 variant

(table 2) (30), implying that the 16189 variants arose independently from in several different maternal

lines.

DISCUSSION

We demonstrated a significant association between the 16189 variant and thinness (low BMI) in both

mothers and their 20 year old offspring, associated with significantly raised placental weight, and

raised placental to birth weight ratio in an Australian birth cohort. Whilst recognising that the power to

detect differences in this study at approx. 5% is too small to provide any proof of an association, the

study nevertheless provides support for the hypothesis and further supports a recent animal study (35)

suggesting mtDNA may be an important genetic contributor to heritable thinness (5). Because

placental surface area and hence weight reflect placental function (21), our study also supports the

suggestion that thinness in adult life may be linked to intrauterine nutrition.

Because ours was a cohort study, it gives a useful estimate of the prevalence of the 16189 variant in the

Caucasian Adelaide population as 8.7%. This is similar to the prevalence found in a different
 Parker et al page 8

Caucasian population (8.8%) (4). Recombination of mtDNA is extremely rare and hence mtDNA

haplotypes reflect maternal origins. Haplotype analysis (Table 2) (30) demonstrated that the 16189

variant had arisen several times independently in this population. Hence, the excess of individuals with

the 16189 variant who were thin at aged 20 was not due to a single mitochondrial founder. This is a

strong finding, because of the advantage that mtDNA haplotyping confers to this study over association

studies of nuclear genes. Since recombination is not a common feature of mitochondrial inheritance

(12), we are able to infer that it is probably the 16189 variant per se, and not co-segregating genetic

factors, that underlie the association with large placenta and thinness in both 20 year olds and their

mothers.

A high placental to birth weight ratio has been observed where mothers smoke, are anaemic (10),

hypoxic due to high altitude(31) or have low carbohydrate intakes early in pregnancy (9) and in babies

that are small for gestational age (13). A unifying hypothesis for these observations would be that the

placenta enlarges to compensates for whatever “fetal stress” results in the thinness. Individuals with

the 16189 variant were thinner, had thinner mothers and higher placental/birthweight ratio than normal,

suggesting that the 16189 variant may act as another such stress. It is not known whether such a trend

is apparent in babies with mtDNA diseases. In the majority of unstressed pregnancies, maternal BMI is

positively correlated with placental/birthweight ratio (15). Other studies have shown that mutations in

maternal and fetal glucokinase genes affecting insulin release may influence the growth of both fetus

(11) and placenta. Hence genes that are important in intrauterine development may also predispose to

diabetes.
 Parker et al page 9

While this study was too small to investigate precisely our previous finding in a UK birth cohort

(n=545) where babies whose postnatal growth subsequently “caught up” and had the 16189 variant

were thinner at birth than those without (4), it confirms that the variant is related to thinness. The

marginally lower insulin levels at age 20 years (figure 1) suggest that thinness may result from low

levels of insulin in utero. This is consistent with the key role that mitochondria play in the

differentiation of adipose tissue and the association of mitochondrial inhibitors with lipodystrophy (22).

It is clear that genes relating to adipocyte metabolism may be important determinants of obesity and

subsequent diabetes (1, 3, 16).

In a previous analysis of the current patient series (8), men who were lighter or shorter as babies were

less insulin sensitive independently of their body mass index or body fat distribution. They also had

higher insulin secretion and increased glucose effectiveness. Overall glucose tolerance, however, did

not correlate with birth size for the whole group. In contrast, the metabolic profiles of individuals with

the 16189 variant in the current study were substantially different from individuals with the lowest

birth weights. While individuals who were small or short at birth and had normal mtDNA were insulin

resistant by aged 20 (8), those with the 16189 variant had normal insulin sensitivity (table 1). Hence

the current data suggest that the association of the 16189 variant with type 2 diabetes is not because the

variant causes low birth weight and thence insulin resistance. Mitochondrial function may also be

important in insulin resistance since there is recent evidence for a coordinated down-regulation of

genes involved in oxidative phosphorylation in skeletal muscle of patients with type 2 diabetes (19,

23). It is thus possible that mild mitochondrial dysfunction, conferred in part by the 16189 variant,

may predispose to type 2 diabetes both at the level of the beta cell, and at the level of glucose disposal

related to ageing. The 16189 variant lies within the control region of mtDNA. A very recent paper
 Parker et al page 10

presents evidence for a novel origin of replication centred on bp 16189 of mtDNA (38). While

previously there was no link to replication, these new data provide a molecular rationale for the

associations described here. We have recently demonstrated that the 16189 variant has a subtle effect

on mitochondrial function, since we have observed a reproducible segregation of the variant in cell

culture, that is characteristic of detrimental mutants (20).

There is however an apparent difference between the effect of the 16189 variant on BMI and on fasting

insulin level in the 20 year olds and in our previous study of 64 year old men, where we found normal

BMI and high fasting insulin levels (28). In neither of these studies was there prospective data

covering the age range where type 2 diabetes normally develops. The progressive deterioration that is

characteristic of mtDNA disease may underlie a change in glucose homeostasis occurring during the

period where we have no data. Furthermore, a decline in mitochondrial function may contribute to the

progressive insulin resistance that is a consequence of aging (24).

In conclusion, this study suggests that the 16189 variant is associated with thinness in young adults and

their mothers. The association with high placental to birth weight ratio suggests some kind of biological

compensation. In contrast with 20 years olds who had been small at birth for other reasons, thinness

associated with the 16189 variant is not accompanied by insulin resistance. As with other causes of small

size in early life, the 16189 variant increases the risk of type 2 diabetes, but apparently by a distinct

mechanism. Further research is required to confirm these hypotheses.

ACKNOWLEDGEMENTS
Financial support for the mitochondrial studies was from The Royal Society, the MRC, the Wellcome Trust
and Diabetes UK. We thank the patients and their physicians for cooperation with the study, particularly Dr
VM Moore. JP is a Royal Society University Research Fellow. We thank Drs A Clark, P Oakeshott, Dr T
 Parker et al page 11

Wallace and Professor D Matthews for useful comments on the manuscript and Professor D Barlow for his
support. All authors declare that they have no competing interests or disclosures.
 Parker et al page 12

Table 1 Comparison of 14 individuals with the 16189 variant with remainder of cohort (n=147) who are normal (wild type)

Wild type: Wild type: Wild type Variant P-values for WT vs. variant
Variable birth weight ≥ 3270g n=99 birth weight < 3270g n=48 n=147 n=14 1 2 3 4
% female 44% --- 58% --- 49% --- 50% --- 0.94 --- --- ---
Maternal age (y) 26.1 5.0 25.7 4.5 25.9 4.8 26.6 3.9 0.37
a
Birth weight (g) 3724 376 2974 220 3480 485 3409 308 0.59 0.60 0.89 ---
b
Ponderal index (g/cm³) 28.5 26.8 to 30.3 25.5 24.1 to 27.4 27.7 25.6 to 29.6 27.9 25.5 to 29.2 0.93 0.93 0.76 0.65
a
Placental weight 611 101 486 63 569 108 629 113 0.051 0.047 0.0025 0.0030
a
Placental weight/ Birth weight 0.164 0.023 0.164 0.024 0.164 0.023 0.185 0.030 0.0024 0.0025 0.019 0.0035
a
Gestation (weeks) 39.7 1.01 39.2 1.30 39.5 1.14 38.9 1.03 0.033 0.032 --- 0.056
b
Maternal BMI (kg/m²) 23.7 21.7 to 25.1 22.4 20.7 to 24.8 23.5 21.4 to 26.6 21.9 20.4 to 22.9 0.013 0.011 0.023 0.019
b
BMI at age 20 (kg/m²) 22.9 20.8 to 25.5 21.5 20.6 to 27.2 22.7 20.8 to 25.6 22.2 21.1 to 23.8 0.019 0.78 0.29 ---
b
Adult biceps skinfold (mm) 5.2 3.6 to 8.2 6.2 3.6 to 10.2 5.4 3.6 to 9.0 4.2 3.0 to 6.2 0.11 0.056 0.067 0.084

Fasting glucose (mmol/l) a 5.5 0.39 5.5 0.39 5.5 0.39 5.4 0.30 0.31x 0.28 0.24 0.35
c
Fasting insulin (pmol/l) 42 25 to 70 48 27 to 87 44 25 to 75 34 21 to 56 0.11 0.095 0.13 0.21
b
Insulin sensitivity 4.70 2.78 to 6.37 4.10 2.27 to 6.68 4.94 2.71 to 6.60 4.08 2.89 to 6.53 0.78 0.40 0.57 0.98
b
Glucose effectiveness 1.69 1.11 to 2.22 1.86 1.50 to 2.71 1.77 1.19 to 2.33 1.66 1.44 to 2.41 0.54 0.52 0.63 0.54

a
mean(SD) b median (IQR) c geometric mean (1SD interval)
P values relate to comparison of the 14 individuals with the 16189 variant with the remaining 147 and are shown in bold where statistically significant at approximately
p<0.05. 1: unadjusted (assuming equal variances except BMI at age 20) 2:adjusted for sex 3: adjusted for sex and gestation 4: adjusted for sex, maternal age, BMI at age 20,
birth weight. To allow for multiple tests, p<0.0038 would be equivalent to p<0.05.
 Parker et al page 13

Table 2: Mitochondrial DNA haplotypes identified in the 14 individuals with the 16189 variant in
the mtDNA hypervariable region 1

Polymorphisms (in region basepairs 15050-16380) (30) Haplotype

T16189C H
T16189C, T16356C H
T16189C, A16183C H
T16189C, A16183C, T16356C, T16362C H
T16189C H
T16189C, A15907G, A16051G, G16129C, A16182C, T16362C U2
T16189C, A15051G, C16111T, G16129, G16145A, A16183C, U2
A16272G, T16362C
T16189C, C16192T*, C16270T, U5
T16189C, C16192T*, C16270T U5
T16189C, C16192T*, C16270T, T16311C, G16336A U5
T16189C, C16270T U5
T16189C, C16176T, A16183C, T16357C, T16449C R
T16189C, G15928A, T16126C, C16294T, T16298C, T16315C, T
C16327T
T16189C, G16129A, C16223T, C16278T X
The 3 individuals marked with an asterix have some mtDNAs with the full 16189 variant but some mtDNAs are
heteroplasmic for C16192T
 Parker et al page 14

REFERENCES

1. Altshuler D, Hirschhorn JN, Klannemark M, Lindgren CM, Vohl MC, Nemesh J, Lane CR, Schaffner SF, Bolk
S, Brewer C, Tuomi T, Gaudet D, Hudson TJ, Daly M, Groop L, and Lander ES. The common PPARgamma Pro12Ala
polymorphism is associated with decreased risk of type 2 diabetes Nat Genet 26: 76-80, 2000.
2. Ashcroft E and Ashcroft S. Mechanisms of insulin secretion. In: Insulin: Molecular Biology to pathology, edited
by Ashcroft F and Ashcroft S. Oxford: OUP, 1992, p. 97-150.
3. Barak Y NM, Ong ES, Jones YZ, Ruiz-Lozano P, Chien KR, Koder A, Evans RM. PPAR gamma is required for
placental, cardiac, and adipose tissue development. Mol Cell 4: 585-595, 1999.
4. Casteels K, Ong K, Phillips D, Bendall H, Pembrey M, The ALSPAC study team, Poulton J, and Dunger D.
Mitochondrial 16189 variant, thinness at birth, and type-2 diabetes. ALSPAC study team. Avon Longitudinal Study of
Pregnancy and Childhood. Lancet 353: 1499-1500, 1999.
5. Costanzo PR and Schiffman SS. Thinness--not obesity--has a genetic component. Neurosci Biobehav Rev 13: 55-
58, 1989.
6. Evans RM, Barish GD, and Wang YX. PPARs and the complex journey to obesity. Nat Med 10: 355-361, 2004.
7. Field P, Ashle N, Morten K, Williams K, Harris D, Clarke A, and Poulton J. In utero exposure to AZT results in
insulin resistance in later life: implications for AIDS prevention in poorly nourished communities. Diabetologia 45 S2:
1217, 2002.
8. Flanagan DE, Moore VM, Godsland IF, Cockington RA, Robinson JS, and Phillips DI. Fetal growth and the
physiological control of glucose tolerance in adults: a minimal model analysis. Am J Physiol Endocrinol Metab 278: E700-
706, 2000.
9. Godfrey K, Robinson S, Barker DJ, Osmond C, and Cox V. Maternal nutrition in early and late pregnancy in
relation to placental and fetal growth. Bmj 312: 410-414., 1996.
10. Godfrey KM, Redman CW, Barker DJ, and Osmond C. The effect of maternal anaemia and iron deficiency on
the ratio of fetal weight to placental weight. Br J Obstet Gynaecol 98: 886-891., 1991.
11. Hattersley AT, Beards F, Ballantyne E, Appleton M, Harvey R, and Ellard S. Mutations in the glucokinase gene
of the fetus result in reduced birth weight. Nat Genet 19: 268-270, 1998.
12. Kraytsberg Y, Schwartz M, Brown TA, Ebralidse K, Kunz WS, Clayton DA, Vissing J, and Khrapko K.
Recombination of human mitochondrial DNA. Science 304: 981, 2004.
13. Lao TT and Wong WM. Placental ratio and intrauterine growth retardation. Br J Obstet Gynaecol 103: 924-926.,
1996.
14. Leary SC and Shoubridge EA. Mitochondrial biogenesis: which part of "NO" do we understand? Bioessays 25:
538-541, 2003.
15. Little RE, Zadorozhnaja TD, Hulchiy OP, Mendel NA, Shkyryak-Nyzhnyk ZA, Chyslovska N, and Gladen BC.
Placental weight and its ratio to birthweight in a Ukrainian city. Early Hum Dev 71: 117-127, 2003.
16. Masud S and Ye S. Effect of the peroxisome proliferator activated receptor-gamma gene Pro12Ala variant on
body mass index: a meta-analysis. J Med Genet 40: 773-780, 2003.
17. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF, and Turner RC. Homeostasis model
assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man.
Diabetologia 28: 412-419, 1985.
18. Montague CT and O'Rahilly S. The perils of portliness: causes and consequences of visceral adiposity. Diabetes
49: 883-888, 2000.
19. Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, Lehar J, Puigserver P, Carlsson E,
Ridderstrale M, Laurila E, Houstis N, Daly MJ, Patterson N, Mesirov JP, Golub TR, Tamayo P, Spiegelman B, Lander
ES, Hirschhorn JN, Altshuler D, and Groop LC. PGC-1alpha-responsive genes involved in oxidative phosphorylation are
coordinately downregulated in human diabetes. Nat Genet 34: 267-273, 2003.
20. Morten K, Jen C, and Poulton J. The 16189 variant of mtDNA in type 2 diabetes: towards a molecular
mechanism. Diabetic Medicine 20 S2: 13, 2003.
21. Owens JA, Falconer J, and Robinson JS. Effect of restriction of placental growth on fetal and utero-placental
metabolism. J Dev Physiol 9: 225-238, 1987.
22. Pace CS, Martin AM, Hammond EL, Mamotte CD, Nolan DA, and Mallal SA. Mitochondrial proliferation, DNA
depletion and adipocyte differentiation in subcutaneous adipose tissue of HIV-positive HAART recipients. Antivir Ther 8:
323-331, 2003.
 Parker et al page 15

23. Patti ME, Butte AJ, Crunkhorn S, Cusi K, Berria R, Kashyap S, Miyazaki Y, Kohane I, Costello M, Saccone R,
Landaker EJ, Goldfine AB, Mun E, DeFronzo R, Finlayson J, Kahn CR, and Mandarino LJ. Coordinated reduction of
genes of oxidative metabolism in humans with insulin resistance and diabetes: Potential role of PGC1 and NRF1. Proc
Natl Acad Sci U S A 100: 8466-8471, 2003.
24. Petersen KF, Befroy D, Dufour S, Dziura J, Ariyan C, Rothman DL, DiPietro L, Cline GW, and Shulman GI.
Mitochondrial dysfunction in the elderly: possible role in insulin resistance. Science 300: 1140-1142, 2003.
25. Petersen KF, Dufour S, Befroy D, Garcia R, and Shulman GI. Impaired mitochondrial activity in the insulin-
resistant offspring of patients with type 2 diabetes. N Engl J Med 350: 664-671, 2004.
26. Poulton J, Luan J, Macaulay V, Hennings S, Mitchell J, and Wareham NJ. Type 2 diabetes is associated with a
common mitochondrial variant: evidence from a population-based case-control study. Hum Mol Genet 11: 1581-1583,
2002.
27. Poulton J, O'Rahilly S, Morten K, and Clark A. Mitochondrial DNA, diabetes and pancreatic pathology in
Kearns-Sayre syndrome. Diabetologia 38: 868-871, 1995.
28. Poulton J, Scott-Brown M, Cooper A, Marchington D, and Phillips D. A common mitochondrial DNA variant is
associated with insulin resistance in adult life. Diabetologia 41: 54-58, 1998.
29. Reardon W, Ross RJ, Sweeney MG, Luxon LM, Pembrey ME, Harding AE, and Trembath RC. Diabetes mellitus
associated with a pathogenic point mutation in mitochondrial DNA. Lancet 340: 1376-1379, 1992.
30. Richards MB, Macaulay VA, Bandelt HJ, and Sykes BC. Phylogeography of mitochondrial DNA in western
Europe. Ann Hum Genet 62: 241-260, 1998.
31. Robinson J, Chidzanja S, Kind K, Lok F, Owens P, and Owens J. Placental control of fetal growth. Reprod Fertil
Dev 7: 333-344, 1995.
32. Silva JP, Kohler M, Graff C, Oldfors A, Magnuson MA, Berggren PO, and Larsson NG. Impaired insulin
secretion and beta-cell loss in tissue-specific knockout mice with mitochondrial diabetes. Nat Genet 26: 336-340, 2000.
33. Soejima A, Inoue K, Takai D, Kaneko M, Ishihara H, Oka Y, and Hayashi J. Mitochondrial DNA is required for
regulation of glucose-stimulated insulin secretion in a mouse pancreatic beta cell line; MIN6. J Biol Chem 271: 26194,
1996.
34. Sue CM, Holmes-Walker DJ, Morris JG, Boyages SC, Crimmins DS, and Byrne E. Mitochondrial gene mutations
and diabetes mellitus [letter]. Lancet 341: 437-438, 1993.
35. Trifunovic A, Wredenberg A, Falkenberg M, Spelbrink JN, Rovio AT, Bruder CE, Bohlooly YM, Gidlof S,
Oldfors A, Wibom R, Tornell J, Jacobs HT, and Larsson NG. Premature ageing in mice expressing defective
mitochondrial DNA polymerase. Nature 429: 417-423, 2004.
36. Vila MR, Gamez J, Solano A, Playan A, Schwartz S, Santorelli FM, Cervera C, Casali C, Montoya J, and
Villarroya F. Uncoupling protein-1 mRNA expression in lipomas from patients bearing pathogenic mitochondrial DNA
mutations. Biochem Biophys Res Commun 278: 800-802, 2000.
37. Welsh N, Paabo S, and Welsh M. Decreased mitochondrial gene expression in isolated islets of rats injected
neonatally with streptozotocin. Diabetologia 34: 626-631, 1991.
38. Yakusawa T, Yang M, Jacobs H, and Holt I. A bidirectional origin of replication maps to the major non-coding
region of human mitochondrial DNA. Molecular Cell in press, 2005.
 Parker et al page 16

Legend to figure 1

Plasma glucose (upper panel) and insulin (lower panel) levels during an IVGTT in subjects at 20 years

old, classified into three groups: variant mtDNA (16189), wild type mtDNA with birth weight ≥3270 g

(WT normal BW), and wild type mtDNA with birth weight < 3270g (WT Low BW). Insulin and

glucose levels during IVGTT were not significantly different in 20 year olds with the 16189 variant

compared with the whole wild type population. When compared with the lower third of birth weight in

the wild type mtDNA group there was a suggestion of lower plasma insulin during the first 20-30

minutes, but this only reached statistical significance at 3 minutes (p=0.0101, controlling for sex and

BMI). Similarly the 3 minute plasma glucose level was significantly lower (p=0.046 controlling for

sex and BMI). These results must be viewed with caution in view of multiple testing and the small

number of individuals with the variant mtDNA. Error bars are +/-2 standard errors.
 Figure 1 upper panel

25
WT normal BW
20 16189
Blood glucose (mM)

WT low BW
15

10

0
-20 0 20 40 60 80 100 120 140 160 180

Time after glucose load (minutes)

 Fig 1 lower panel

800
700 WT normal BW
WT low BW
600
Blood insulin (pmol)

16189
500
400
300
200
100
0
-20 0 20 40 60 80 100 120 140 160 180

Time after glucose load (minutes)