A male excess in Sardinian type 1 diabetic cases has previously been reported and was largely restricted to those patients carrying the HLA-DR3/nonDR4 genotype. In the present study, we have measured the male- to-female (M:F) ratio in a sample set of 542 newly collected, early-onset type 1 diabetic Sardinian patients. This data not only confirm the excess of male type 1 diabetic patients overall (M:F ratio = 1.3, P = 3.9 × 10−3) but also that the bias in male incidence is largely confined to patients with the DR3/nonDR4 genotype (M:F ratio = 1.6, P = 2.0 × 10−4). These sex effects could be due to a role for allelic variation of the Y chromosome in the susceptibility to type 1 diabetes, but to date this chromosome has not been evaluated in type 1 diabetes. We, therefore, established the frequencies of the various chromosome Y lineages and haplotypes in 325 Sardinian male patients, which included 180 cases with the DR3/nonDR4 genotype, and 366 Sardinian male control subjects. Our results do not support a significant involvement of the Y chromosome in DR3/nonDR4 type 1 diabetic cases nor in early-onset type 1 diabetes as a whole. Other explanations, such as X chromosome-linked inheritance, are thus required for the male bias in incidence in type 1 diabetes in Sardinia.

Type 1 diabetes is the only common autoimmune trait that does not exhibit a strong female excess in the patients. A male excess has been reported in specific subgroups of type 1 diabetic patients, defined by the population of origin, the age of onset of the disease, and the genotypes at the major disease locus, HLA/IDDM1. In early-onset type 1 diabetes, with a diagnosis under the age of 15 years, an increased male-to-female (M:F) ratio was observed in the patients from the high type 1 diabetes risk population of Sardinia (M:F ratio = 1.5) (1). To a lesser extent, the bias is also observed in other high incidence European populations (1). It has been reported that in Sardinia the male excess is largely restricted to those patients carrying the DR3/nonDR4 genotype at the major disease locus, IDDM1. No significant sex effects were observed in the DR4/nonDR3, DR3/4, and nonDR3/nonDR4 categories (2). A male bias is consistently observed in patients diagnosed between the ages of 15 and 40 years from across Europe, but this appears to be independent of HLA type (3). Overall, these observations are consistent with an involvement of the sex chromosomes in type 1 diabetes. Suggestive evidence of linkage of type 1 diabetes to chromosome Xp22-p11 has been reported in families with DR3/nonDR4-affected sib-pairs (2). Nevertheless, it is possible that chromosome Y might account, at least in part, for the above described sex effects in type 1 diabetes. Moreover, some nonallelic homologues of genes located on chromosome X are contained in the nonrecombinant portion of chromosome Y (NRY). For instance, chromosome Y encodes a minor histocompatibility antigen (UTY) homologue of (UTX) (4), which is located in the chromosome X interval (Xp22-p11) that showed linkage to type 1 diabetes (2). Despite these observations, chromosome Y is the only portion of the human genome that has never been scanned for its involvement in type 1 diabetes. Exclusion mapping of the Y chromosome for disease association can be carried out by using a relatively low number of markers by virtue of the absence of crossing-over in the NRY. In the present report, we have tested the possibility that the Y chromosome is responsible for the high M:F ratio using a large male case-control sample set from Sardinia.

We have first determined the M:F ratio bias in a sample set of 542 Sardinian patients, diagnosed under age 15 years, analyzed for the first time in the present study (Table 1). The analysis of these newly reported cases provides a clear confirmation that in Sardinian early-onset patients there is a male excess (M:F ratio = 1.3, one-tailed P = 3.9 × 10−3) and that the bias in male incidence is almost exclusively restricted to patients with the DR3/nonDR4 genotype (M:F ratio = 1.6, one-tailed P = 2.0 × 10−4) (Table 1). In particular in this new collection of cases, we found significant heterogeneity, evaluated using a 2 × 2 contingency table and tested by a χ2 test, in the M:F ratios between the DR3/nonDR4 and DR4/nonDR3 genotypes (P = 3.3 × 10−3). These results are similar to those we reported previously in an independent sample set of 325 Sardinian type 1 diabetic subjects (2). We also analyzed the M:F ratio in 471, fully HLA-typed, unaffected siblings of type 1 diabetic patients. We observed an overall M:F ratio of 0.9 with a similar value also in the DR3/nonDR4 category (98 males and 109 females, M:F ratio = 0.9). Thus, there was significant heterogeneity in the M:F ratios in the DR3/nonDR4 category between the affected patients and their unaffected siblings (P = 1.1 × 10−4).

Owing to the strong male excess in patients and the high frequency of HLA-DR3, Sardinia offers a special opportunity to assess if the Y chromosome is involved in the inheritance of type 1 diabetes. We, therefore, evaluated the association of the Y chromosome with type 1 diabetes by using a male case-control sample set from Sardinia of 325 unrelated early-onset male and 366 unrelated male control subjects. The main chromosome Y lineages and haplotypes were established (see research design and methods) and their frequencies contrasted in patients and control subjects (Table 2). The haplotypes shown in Table 2 define 97.3% of the chromosome Y variability present in the general Sardinian population, while the remaining variation is accounted for by a combination of very rare lineages and haplotypes (D.C., F.C., unpublished results). There is no significant association of any chromosome Y marker with type 1 diabetes in this dataset. The Eu10/Eu12 was under-represented in the patients when compared with the control subjects (P = 2.8 × 10−2), but the difference was no longer significant at the 5% level after correction for number of haplotypes compared (n = 8) (Table 2). Importantly, we did not observe any significant difference in the frequencies of the various chromosome Y markers also when the data were stratified according to the genotypes at the HLA-DRB1 and -DQB1 loci, including the DR3/nonDR4 subgroup (Table 2 and data not shown). The power to detect gene effects with an odds ratio ≥2.5 was >95% at P = 5 × 10−3 for all the chromosome Y haplotypes with the exception of Eu7, for which we had 87.9% power to detect such a gene effect (Table 2). For such a gene effect, even considering the sample size of 180 patients carrying the DR3/nonDR4 genotype, we obtain power >95% for all the chromosome Y haplotypes with the exception of Eu10/12 and Eu7, for which we had 86 and 67.4% power, respectively, to detect association at P = 5 × 10−3 (Table 2).

Finally, we evaluated 11 father-affected son-pairs. We reasoned that even if the Y chromosome was contributing to a small portion of the inheritance of type 1 diabetes, these families were most likely to contain a disease-associated variant. However, we found that these father-son pairs carried four different chromosome Y lineages: Eu4 (three pairs), Eu8 (six pairs), Eu9 (one pair), and Eu18 (one pair), the distribution of which did not significantly differ from that observed in the general population. These results contrast with the existence of Sardinian founder mutations involved in monogenic disorders such as Thalassemia (5), Wilson disease (6), and APECED (autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy) (7) and do not support a founder Sardinian chromosome Y polymorphism contributing to the inheritance of type 1 diabetes in these families. Overall, from our data we can conclude that variation at the NRY is unlikely to contribute in a significant way to the inheritance of type 1 diabetes.

Other explanations are, therefore, required to explain the sex effects observed in this common multifactorial disease. A role for chromosome X is suggested by evidence of linkage of type 1 diabetes to Xp22.11 detected in affected sib-pair families from the U.K. and the U.S. (2,8) and, more recently, in families largely of Scandinavian descent (9). However, the explanation for the sex effects observed in type 1 diabetes could be very complex, because the M:F ratio varies in different countries; in general, with more females affected in countries with lower incidence (10). Either the allele frequency of the putative chromosome X causal variant(s) differs among these distinct populations or the population frequencies of interacting loci at other chromosome locations vary (for example, DR3/nonDR4 genotype frequencies show marked differences across distinct ethnic groups). Additionally but not exclusively, interacting environmental factors may vary in prevalence among countries and influence the penetrance of the putative chromosome X variant in the etiology of type 1 diabetes.

Finally, we excluded the possibility that the male excess in type 1 diabetes that we observed in Sardinian children is due to variation in the levels of the sex hormones during puberty. The initial stages of sexual maturity begin after 9 years of age in over 97% of boys (11) while the age of onset of type 1 diabetes in the Sardinian patients averages <8 years (reasearch design and methods). However, since some of our cases were>9 years of age at diagnosis (but <15 years), we also determined the M:F ratio in only those cases who were ≤8 years at disease onset (n = 446, including cases from our first study). In these patients the male bias was even more pronounced, with 265 males and 181 females overall (M:F ratio = 1.5), and 155 males and 77 females in the DR3/nonDR4 category (M:F ratio = 2). The results presented here provide further justification for the identification of genes on chromosome X that contribute to type 1 diabetes inheritance.

Sample selection.

The total number of type 1 diabetes cases studied from Sardinia was 867, of which 325 were studied previously (2) and 542 were new and unrelated to the previous cases. The average age of the patients at disease onset was 7.8 years, SD ±3.8 years (minimum 0.5 maximun 14); 87.1% of the patients were from the province of Cagliari, 4.2% were from the province of Sassari, 4.3% were from the province of Nuoro, and 4.3% were from the province of Oristano. The involvement of chromosome Y has been examined in a sample set consisting of 325 unrelated male type 1 diabetic patients (selected from the total sample set of 867 patients) and 366 unrelated male control subjects. The average age of onset of these male patients was 7.9 years, SD ±3.9 years (minimum 0.5, maximum 14). The 366 Sardinian control samples were collected from 155 healthy adult male blood donors and 211 newborns that were referred to our center for neonatal screening programs. To restrict the analysis to individuals whose descent was from Sardinia, the surnames of both the patients and control subjects were considered by using the software gens (www.gens.labo.net).

The DNA from the type 1 diabetic patients and the blood donors was prepared from peripheral blood and purified by using a standard salting-out procedure. The DNA from the newborn samples was extracted from dried blood spots of the Guthrie Cards by using the Chelex method (Chelex 100; Bio-Rad, Hercules, CA) (12).

Genotyping and statistical analysis of the data.

The HLA class II typing has been performed as previously described (13). The M:F distribution and M:F ratios were examined in the patients carrying the four main HLA genotype/categories (DR3/nonDR4, DR3/DR4, DR4/nonDR3, and nonDR3/nonDR4) as well as in the patients unstratified by HLA genotype. The P values were one-tailed based on our prior observations (2) and computed using a 2 × 1 contingency table versus a null hypothesis of equal distribution of males and females in each category (Table 1).

Male samples of 325 type 1 diabetic patients and 366 control subjects were then typed for nine chromosome Y diallelic polymorphisms and one Alu insertion (Yap) located in the NRY selected for their European-specific distribution sample (14). We used a step-wise genotyping approach based on the relative frequencies of the various chromosome Y markers obtained from a previous analysis on a smaller Sardinian sample (14). Since phylogenetically assessed, these polymorphisms were classified by Underhill et al. (15) in order of descent and called M1, M9, M26, M35, M89, M170, M172, M173, and M201. M1 (YAP) was typed by using the method of Hammer and Horai (16). M9 and M35, were typed by restriction fragment-length polymorphism using HinfI and BrsI, respectively (15); the M89 and M170 single-nucleotide polymorphisms with an amplification refractory mutation system PCR (17). Dot Blot analysis was used for M26 and M173. M201 and M172 were typed by using the transgenomic system for denaturing high-performance liquid chromatography analysis (18). Primers and probes used in this study are shown in Table 3.

The distribution of the patient and control lineages and haplotypes, determined with a gene counting procedure, was then arranged in a 2 × 2 contingency table and tested by Fisher’s exact or Pearson’s χ2 test and the ORs were computed. The statistical power of our sample set has been computed considering the individual frequencies in the general population of the various lineages and haplotypes based on standard epidemiological measures applied to 2 × 2 contingency tables. We based our power calculations on a gene effect with OR = 2.5, a value comparable to the OR observed for the class I VNTR allele at INS/IDDM2 on chromosome 11p15.

TABLE 1

M:F ratio in 542 early-onset type 1 diabetic patients from Sardinia according to the HLA/IDDM1 genotypes

Total
DR3/nonDR4
DR3/DR4
DR4/nonDR3
nonDR3/nonDR4
nP < 0.05nP < 0.05nP < 0.05nP < 0.05nP < 0.05
Male 302  158  107  29  8  
Female 240  101  94  41  4  
Sum 542  259  201  70  12  
M:F ratio 1.3 3.9E–03 1.6 2.0E–04 1.1  0.7  2  
Total
DR3/nonDR4
DR3/DR4
DR4/nonDR3
nonDR3/nonDR4
nP < 0.05nP < 0.05nP < 0.05nP < 0.05nP < 0.05
Male 302  158  107  29  8  
Female 240  101  94  41  4  
Sum 542  259  201  70  12  
M:F ratio 1.3 3.9E–03 1.6 2.0E–04 1.1  0.7  2  
View Large
TABLE 2

The distribution of chromosome Y haplotypes in 325 Sardinian early-onset type 1 diabetic patients and in 366 ethnically matched control subjects

Haplotypes†Patients
Total
DR3/nonDR4
Control subjects
OR95% CIPower*
n%n%n%TotalDR3/nonDR4
Eu4 22 6.8 12 6.7 33 9.0 0.7 0.4–1.3 99.9% 98.6% 
Eu10/12‡ 6 1.8 1 0.6 18 4.9 0.4 0.1–0.9 97.0% 86.0% 
Eu7 9 2.8 4 2.2 12 3.3 0.8 0.3–2.0 87.9% 67.4% 
Eu8 131 40.3 74 41.1 141 38.5 1.1 0.8–1.5 99.9% 99.9% 
Eu9 34 10.5 24 13.3 34 9.3 1.1 0.7–1.9 99.9% 98.8% 
Eu11 34 10.5 19 10.6 46 12.6 0.8 0.5–1.3 99.9% 99.9% 
Eu18/Eu19 81 24.9 41 22.8 72 19.7 1.4 0.9–1.9 99.9% 99.9% 
Others 8 2.5 5 2.8 10 2.7 0.9 0.4–2.3   
Sum 325  180  366      
Haplotypes†Patients
Total
DR3/nonDR4
Control subjects
OR95% CIPower*
n%n%n%TotalDR3/nonDR4
Eu4 22 6.8 12 6.7 33 9.0 0.7 0.4–1.3 99.9% 98.6% 
Eu10/12‡ 6 1.8 1 0.6 18 4.9 0.4 0.1–0.9 97.0% 86.0% 
Eu7 9 2.8 4 2.2 12 3.3 0.8 0.3–2.0 87.9% 67.4% 
Eu8 131 40.3 74 41.1 141 38.5 1.1 0.8–1.5 99.9% 99.9% 
Eu9 34 10.5 24 13.3 34 9.3 1.1 0.7–1.9 99.9% 98.8% 
Eu11 34 10.5 19 10.6 46 12.6 0.8 0.5–1.3 99.9% 99.9% 
Eu18/Eu19 81 24.9 41 22.8 72 19.7 1.4 0.9–1.9 99.9% 99.9% 
Others 8 2.5 5 2.8 10 2.7 0.9 0.4–2.3   
Sum 325  180  366      

OR and 95% CI have been calculated by comparing the frequencies of the various haplotypes in the total patients (N = 325) and control subjects (N = 366).

*

Statistical power has been computed as the probability to detect a gene effect with an OR = 2.5 at a P value = 5 × 10−3 considering the individual frequencies in the general population of the various haplotypes. In computing the statistical power we considered both the sample size of 325 patients not stratified by the HLA genotypes and the sample size of 180 patients carrying the DR3/nonDR4 genotype.

†

Only haplotypes with a frequency of >2% in patients or control subjects are shown in this table.

‡

Individuals positive for M89 and negative for M170, M172, M201, and M9.

View Large
TABLE 3

List of primers and probes redesigned for this study

PolymorphismPrimers and probes
M26 5′ CCAGTGGTAAAGTTTTATTACAATTTTTTT 
 3′ TTCACAGTAAGCAGGCAATCCAATGTCAGG 
 M26-G GAATTATCGCAGAGAACAC 
 M26-A AGAATTATCACAGAGAACA 
M35 5′ TAAGCCTAAAGAGCAGTCAGAGTAGAATGC 
 3′ AGAGGGAGCAATGAGGACATTATTCTCTAA 
M89 5′ AGAAGCAGATTGATGTCCCACTTAAAGAAG 
 3′ TCCAGTTAGGAGATCCCCTCATGATCCCAC 
 FOR-C TCTCTTCCTAAGGTTATGTACAAAAATATC 
 REV-A GTAGCTGCAACTCAGGCAAAGTGAGAGGTT 
M170 5′ TGCTTCACACAAATGCGTTTCAAATAGTAA 
 3′ CCAATTACTTTCAACATTTAAGACCACACA 
 FOR-A CTATTTTATTTACTTAAAAATCATTGTCCA 
 REV-C CAGTGAGACACAACCCACACTGAAAAACAG 
M172* 5′ TTGAAGTTACTTTTATAATCTAATGCTTAA 
 3′ ATAATTTATTACTTTACAGTCACAGTGGAT 
 5′* ATGCCCCAGATGAGCATGAG 
 3′* CAGCCAGGTACAGAGAAAGTTTG 
M173 5′ AAGAAATGTTGAACTGAAAGTTGATGCCAC 
 3′ AGGTGTATCTGGCATCCGTTAGAAAAGTCA 
 M173-C GCATTTAGAACCCTTTGTCATCTG 
 M173-A GCATTTAGAACACTTTGTCATCT 
M201 5′ ATAGTTACTACTTGAGTTACTATATTAGTG 
 3′ AAACGTCAAACAGTACCTGTATTTCCTTAG 
PolymorphismPrimers and probes
M26 5′ CCAGTGGTAAAGTTTTATTACAATTTTTTT 
 3′ TTCACAGTAAGCAGGCAATCCAATGTCAGG 
 M26-G GAATTATCGCAGAGAACAC 
 M26-A AGAATTATCACAGAGAACA 
M35 5′ TAAGCCTAAAGAGCAGTCAGAGTAGAATGC 
 3′ AGAGGGAGCAATGAGGACATTATTCTCTAA 
M89 5′ AGAAGCAGATTGATGTCCCACTTAAAGAAG 
 3′ TCCAGTTAGGAGATCCCCTCATGATCCCAC 
 FOR-C TCTCTTCCTAAGGTTATGTACAAAAATATC 
 REV-A GTAGCTGCAACTCAGGCAAAGTGAGAGGTT 
M170 5′ TGCTTCACACAAATGCGTTTCAAATAGTAA 
 3′ CCAATTACTTTCAACATTTAAGACCACACA 
 FOR-A CTATTTTATTTACTTAAAAATCATTGTCCA 
 REV-C CAGTGAGACACAACCCACACTGAAAAACAG 
M172* 5′ TTGAAGTTACTTTTATAATCTAATGCTTAA 
 3′ ATAATTTATTACTTTACAGTCACAGTGGAT 
 5′* ATGCCCCAGATGAGCATGAG 
 3′* CAGCCAGGTACAGAGAAAGTTTG 
M173 5′ AAGAAATGTTGAACTGAAAGTTGATGCCAC 
 3′ AGGTGTATCTGGCATCCGTTAGAAAAGTCA 
 M173-C GCATTTAGAACCCTTTGTCATCTG 
 M173-A GCATTTAGAACACTTTGTCATCT 
M201 5′ ATAGTTACTACTTGAGTTACTATATTAGTG 
 3′ AAACGTCAAACAGTACCTGTATTTCCTTAG 
*

Two different primers for the M172 typing were needed and designed to obtain a smaller fragment for those DNA samples extracted from Guthrie Cards. Probes used to type M26 and M173 with the Dot Blot analysis are indicated in italics. The polymorphisms are indicated in bold text.

View Large

F.C. and J.A.T. are recipients of a Wellcome Trust Biomedical Research Collaboration Grant. We thank the Italian Telethon, the Juvenile Diabetes Research Foundation, the Regione Autonoma Sardegna Assessorato Sanita’, and the Wellcome Trust for financial support.

Thanks to Antonio Cao for continuous support and encouragement. We also thank Michael Whalen, Mauro Congia, M. Antonietta Asunis, and Mario Silvetti for help and advice; Heather Cordell for statistical advice; Margi Chessa and Rossella Ricciardi for help in collecting the Sardinian type 1 diabetic family and clinical information; and Maria Melis and Antonella Deidda for taking the blood of the patients and their relatives.

1
Karvonen M, Viik-Kajander M, Moltchanova E, Libman I, LaPorte R, Tuomilehto J: Incidence of childhood type 1 diabetes worldwide: Diabetes Mondiale (DiaMond) Project Group.
Diabetes Care
 
23
:
1516
–1526,
2000
2
Cucca F, Goy JV, Kawaguchi Y, Esposito L, Merriman ME, Wilson AJ, Cordell HJ, Bain SC, Todd JA: A male-female bias in type 1 diabetes and linkage to chromosome Xp in MHC HLA-DR3-positive patients.
Nat Genet
 
19
:
301
–302
1998
3
Weets I, Van Autreve J, Van der Auwera BJ, Schuit FC, Du Caju MV, Decochez K, De Leeuw IH, Keymeulen B, Mathieu C, Rottiers R, Dorchy H, Quartier E, Gorus FK: Male-to-female excess in diabetes diagnosed in early adulthood is not specific for the immune-mediated form nor is it HLA-DQ restricted: possible relation to increased body mass index.
Diabetologia
 
44
:
40
–47,
2001
4
Vogt MH, Goulmy E, Kloosterboer FM, Blokland E, de Paus RA, Willemze R, Falkenburg JH: UTY gene codes for an HLA-B60-restricted human male-specific minor histocompatibility antigen involved in stem cell graft rejection: characterization of the critical polymorphic amino acid residues for T-cell recognition.
Blood
 
96
:
3126
–3132,
2000
5
Rosatelli MC, Dozy A, Faa V, Meloni A, Sardu R, Saba L, Kan YW, Cao A: Molecular characterization of beta-thalassemia in the Sardinian population.
Am J Hum Genet
 
50
:
422
–426,
1992
6
Loudianos G, Dessi V, Lovicu M, Angius A, Figus A, Lilliu F, De Virgiliis S, Nurchi AM, Deplano A, Moi P, Pirastu M, Cao A: Molecular characterization of wilson disease in the Sardinian population-evidence of a founder effect.
Hum Mutat
 
14
:
294
–303,
1999
7
Rosatelli MC, Meloni A, Devoto M, Cao A, Scott HS, Peterson P, Heino M, Krohn KJ, Nagamine K, Kudoh J, Shimizu N, Antonarakis SE: A common mutation in Sardinian autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy patients.
Hum Genet
 
103
:
428
–434,
1998
8
Cordell HJ, Kawaguchi Y, Todd JA, Farrall M: An extension of the maximum lod score method to X-linked loci.
Ann Hum Genet
 
59
:
435
–449,
1995
9
Nerup J, Pociot F: A genomewide scan for type 1-diabetes susceptibility in Scandinavian families: identification of new loci with evidence of interactions.
Am J Hum Genet
 
69
:
1301
–1313,
2001
10
Karvonen M, Pitkaniemi M, Pitkaniemi J, Kohtamaki K, Tajima N, Tuomilehto J: Sex difference in the incidence of insulin-dependent diabetes mellitus: an analysis of the recent epidemiological data. World Health Organization DIAMOND Project Group.
Diabetes Metab Rev
 
13
:
275
–291,
1997
11
Tanner JM, Davies PS: Clinical longitudinal standards for height and height velocity for North American children.
J Pediatr
 
107
:
317
–329,
1985
12
Walsh PS, Metzger DA, Higuchi R: Chelex 100 as a medium for simple extraction of DNA for PCR-based typing from forensic material.
Biotechniques
 
10
:
506
–513,
1991
13
Cucca F, Lampis R, Frau F, Macis D, Angius E, Masile P, Chessa M, Frongia P, Silvetti M, Cao A, et al: The distribution of DR4 haplotypes in Sardinia suggests a primary association of type I diabetes with DRB1 and DQB1 loci.
Hum Immunol
 
43
:
301
–308,
1995
14
Semino O, Passarino G, Oefner PJ, Lin AA, Arbuzova S, Beckman LE, De Benedictis G, Francalacci P, Kouvatsi A, Limborska S, Marcikiae M, Mika A, Mika B, Primorac D, Santachiara-Benerecetti AS, Cavalli-Sforza LL, Underhill PA: The genetic legacy of Paleolithic Homo sapiens sapiens in extant Europeans: a Y chromosome perspective.
Science
 
290
:
1155
–1159,
2000
15
Underhill PA, Shen P, Lin AA, Jin L, Passarino G, Yang WH, Kauffman E, Bonne-Tamir B, Bertranpetit J, Francalacci P, Ibrahim M, Jenkins T, Kidd JR, Mehdi SQ, Seielstad MT, Wells RS, Piazza A, Davis RW, Feldman MW, Cavalli-Sforza LL, Oefner PJ: Y chromosome sequence variation and the history of human populations.
Nat Genet
 
26
:
358
–361,
2000
16
Hammer MF, Spurdle AB, Karafet T, Bonner MR, Wood ET, Novelletto A, Malaspina P, Mitchell RJ, Horai S, Jenkins T, Zegura SL: The geographic distribution of human Y chromosome variation.
Genetics
 
145
:
787
–805,
1997
17
Powis SH, Tonks S, Mockridge I, Kelly AP, Bodmer JG, Trowsdale J: Alleles and haplotypes of the MHC-encoded ABC transporters TAP1 and TAP2.
Immunogenetics
 
37
:
373
–380,
1993
18
Underhill PA, Jin L, Lin AA, Mehdi SQ, Jenkins T, Vollrath D, Davis RW, Cavalli-Sforza LL, Oefner PJ: Detection of numerous Y chromosome biallelic polymorphisms by denaturing high-performance liquid chromatography.
Genome Res
 
7
:
996
–1005,
1997

Address correspondence and reprint requests to Francesco Cucca, Dipartimento di Scienze Biomediche e Biotecnologie, University of Cagliari, Via Jenner, Cagliari 09121, Italy. E-mail: [email protected].

Received for publication 10 June 2002 and accepted in revised form 26 August 2002.

D.C. and L.M. contributed equally to this article.

M:F, male-to-female; NRY, nonrecombinant portion of chromosome Y.

DIABETES
2002