Association of tumor necrosis factor alpha gene -G308A polymorphism with schizophrenia

Schizophrenia Research 65 (2003) 19 – 25 www.elsevier.com/locate/schres

Association of tumor necrosis factor alpha gene -G308A polymorphism with schizophrenia Sibylle G. Schwab a, Stephanie Mondabon a, Michael Knapp b, Margot Albus c, Joachim Hallmayer d, Margitta Borrmann-Hassenbach c, Matyas Trixler e, Magdalena Groh f, Thomas G. Schulze f,g, Marcella Rietschel f,h, Bernard Lerer i, Wolfgang Maier f, Dieter B. Wildenauer a,* a

Molecular Genetics Laboratory, Department of Psychiatry, University of Bonn, Wilhelmstr. 31, D-53111 Bonn, Germany b Department of Medical Statistics, University of Bonn, D-53105 Bonn, Germany c Mental State Hospital, D-85529 Haar, Germany d Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Palo Alto, CA, USA e Department of Psychiatry, University of Pecs, Pecs, Hungary f Department of Psychiatry, University of Bonn, D-53105 Bonn, Germany g Department of Psychiatry, The University of Chicago, Chicago, IL, USA h Zentralinstitut fu¨r Seelische Gesundheit, D-68072 Mannheim, Germany i Department of Psychiatry, Hadassah-Hebrew University Medical Center, 91120 Jerusalem, Israel Received 3 December 2002; accepted 24 December 2002

Abstract Background: Tumor necrosis factor alpha (TNFa), a cytokine involved in inflammatory processes, has been implicated in the pathophysiology of schizophrenia. The chromosomal location in the major histocompatibility complex (MHC) region on 6p21.1 – 21.3, a region with evidence for linkage, suggests a role in susceptibility to schizophrenia. Association of the minor (A) allele of the -G308A TNFa gene polymorphism with schizophrenia has been reported [Mol. Psychiatry 6 (2001) 79]. Methods: Association of the -G308A TNFa gene and the lymphotoxin alpha (LTa) + A252G gene polymorphisms with schizophrenia was studied in 79 sib pair families with linkage in the MHC region and in 128 trio families using the transmission disequilibrium test (TDT). Results: Weak association of the common G allele was detected for TNFa -G308A in both samples independently with borderline significance in the sib pair families (0.064) and with a nominally significant value of P = 0.022 in the trio families. Combining both samples produced P = 0.003, while LTa + A252G, located approximately 2 – 3 kb distally, revealed P = 0.03 and the two locus haplotype yielded a P value of 0.001. Conclusion: Our data suggests association of the common G allele of the -G308A TNFa gene polymorphism with schizophrenia in a sample of 207 families. However, linkage disequilibrium with a different allele of the TNFa gene or another gene in the MHC region cannot be excluded. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Cytokines; TNFa; LTa; Schizophrenia; Association; Linkage disequilibrium

* Corresponding author. Tel.: +49-228-287-2352; fax: +49-228-287-2617. E-mail address: [email protected] (D.B. Wildenauer). 0920-9964/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0920-9964(02)00534-0

20

S.G. Schwab et al. / Schizophrenia Research 65 (2003) 19–25

1. Introduction A role of the immune system in pathophysiology of schizophrenia has been discussed for many years. Association of specific human leukocyte antigens may mediate autoimmune response resulting in alterations in neuronal structures (Wright et al., 2001). There is also evidence that altered levels of cytokines, a class of proteins or peptides involved in signaling between cells during immune response, may contribute to the development of schizophrenia. Proinflammatory cytokines like interleukin (IL)-1, 2, -6, and TNFa have been found to be increased in plasma or cerebrospinal fluid of schizophrenic patients (Licinio et al., 1993; Ganguli et al., 1994; Maes et al., 1995, 2000; Monteleone et al., 1997; Naudin et al., 1997; Lin et al., 1998; Wright et al., 2001). It has been shown that cytokines are not restricted in their action to peripheral sites. They can act directly on the central nervous system by crossing the blood – brain barrier and indirectly by signaling the brain via peripheral nerves or by secondary messengers like NO (nitric oxide) or prostaglandins (Licinio and Wong, 1997). Moreover, genes for several cytokines are constitutively expressed in the brain (Wong and Licinio, 1994). Evidence for the release of TNFa from glial cells in cultured hippocampal neurons and in hippocampal slices has been reported recently by Beattie et al. (2002). These authors show that the release of TNFa enhances synaptic efficacy by increasing surface expression of AMPA receptors, while preventing the actions of endogenous TNFa has the opposite effects. This suggests a role in synaptic plasticity. Thus, it is conceivable that disturbances in the level of TNFa may affect functioning of the brain. Clinical findings have been complemented by studies of association of DNA sequence polymorphisms in the genes for the interleukin-1 gene complex (Katila et al., 1999) as well as in the gene for TNFa (Boin et al., 2001) with schizophrenia. Boin et al. (2001) analyzed the distribution of a G/A transition polymorphism in the promoter region of TNFa (-G308A) in 84 unrelated schizophrenic patients and in 138 unrelated healthy controls. Frequency of the A (TNF2) allele was increased in schizophrenic patients as compared to the control

group ( P = 0.0042). The higher frequency could have functional consequences since association of the A allele with higher TNFa production has been reported (Wilson et al., 1997). They found that the in vitro transcription rate for the A allele is seven times higher than for the G allele. Based on this finding, Boin et al. (2001) concluded that abnormal TNFa production may enhance the risk of developing schizophrenia. However, the enhanced transcription rate for the A allele could not be confirmed by Brinkman et al. (1995). These authors analyzed both allelic forms (TNFa(-308G) and TNFa(-308A)) of the TNFa enhancer/promoter region (
598/ + 108) in a transient transfection system, using chloramphenicol acetyltransferase (CAT) as reporter gene and could not detect differences in the level of inducible reporter gene expression between TNF(-308G)/CAT and TNF(-308A)/ CAT constructs. We were intrigued by the fact that the gene for TNFa is located within the MHC region at chromosome 6p21.31, a region where we have observed potential linkage in our recent genomewide survey for schizophrenia susceptibility genes in 71 sib pair families with schizophrenia (Schwab et al., 2000). In this study, multipoint-affected sib pair analysis produced a nonparametric LOD score of 3.4 ( P = 0.0004) peaking around marker DQB CAR, which is located approximately 900 kb proximal of the TNFa gene (Schwab et al., 2002). Since complete parental genotype information is available, we decided to test the -G308A polymorphism for association/linkage disequilibrium in the sib pair sample and in a sample with 128 offspring/parents (trios) using the TDT. In addition to the -G308A polymorphism in the TNFa gene, we tested a polymorphism representing a G to A transition in the gene for lymphotoxin a (LTa) (formerly known as TNFh). The polymorphism is located in the first intron at nucleotide position + 252 and has been reported to affect expression of the gene (Messer et al., 1991). The TNFa and LTa genes are arranged in tandem in the class III region of MHC and are separated by a stretch of approximately 2000 base pairs. Linkage disequilibrium between the two genes has been demonstrated (Pociot et al., 1993) enabling haplotype formation for association studies.

S.G. Schwab et al. / Schizophrenia Research 65 (2003) 19–25

2. Material and methods 2.1. Families All individuals were interviewed using the Schedule for Affective Disorder Schizophrenia-Lifetime Version (SADS-L) or the Structured Clinical Interview for DSM-III-R Disorders (SCID-II) (Fyer et al., 1985). Case records were evaluated by OPCRIT (McGuffin et al., 1991). The complete interview form, family history information, and the medical record of each individual were reviewed by an independent psychiatrist without prior knowledge of the family relationship to the index patient, or of the morbidity in the family. Composition of the sib pair sample is shown in Table 1. Seventy-two families were from Germany (Haar, Mainz, Bonn), one from Hungary (Pecs) and six were from Israel. The families from Germany were of European ancestry, the Israeli families consisted of four families of non-Ashkenazi, one of Ashkenazi, and one of Arab origin. The 172 affected offspring of the sib pair sample consisted of 160 individuals with core diagnosis schizophrenia, and 19 individuals with diagnosis schizoaffective disorder, schizophrenic type according to the Research Diagnostic Criteria (RDC, Spitzer et al., 1978). The sample of 128 trios (Table 1) consisted of 124 families with one affected offspring with schizophrenia and four families with one affected offspring with schizoaffective disorder, schizophrenic

Table 1 Composition of the family samples (for each family parental genotype available) Sample

Number of families

Number of affected offspring

Sib pairs

68 with pairsa 9 with tripletsb 1 with quadruplets 1 with quintuplets 79 128 207

136 27 4 5 172 128 300

Total Trios Sib pairs + trios combined

a Four sib pairs with one parent and additional unaffected siblings. b Four triplets with one parent and additional unaffected siblings.

21

type according to RDC. The families were of European origin having a family history with psychiatric disorders, i.e. presence of at least one additional first- or second-degree member affected with either schizophrenia or bipolar or recurrent unipolar disorder. Seventy-four families were collected in Haar, 45 in Bonn, and 9 in Pecs (Hungary) as an independent sample for investigation of family-based association/linkage disequilibrium (Spielman et al., 1993). 2.2. Genotyping DNA was isolated from whole blood or permanent cell lines derived from Epstein – Barr virus transformed lymphocytes using a Qiagen Blood- and CellCulture DNA kit. Polymerase chain reaction was performed using 25 ng DNA for the TNFa -G308A polymorphism using the forward oligonucleotide primer 5V AGGCAATAGGTTTTGAGGGCCAT 3V and the reversed primer 5V TCCTCCCTGCTCCGATTCCG 3V as described by Wilson et al. (1992) and for the LTa + A252G polymorphism using the forward oligonucleotid primer 5V CCGTGCTTCGTGCTTTGGACTA 3V and the reversed primer 5V AGAGCTGGTGGGGACATGTCTG 3V as described by Moffat and Cookson (1997). Cycling conditions were 5-min denaturation at 95 jC, followed by 31 cycles, each cycle consisting of 20-s denaturation at 95 jC, 30-s annealing at 55 jC, and 40-s elongation at 72 jC. A final extension step was included consisting of 5 min at 72 jC. Digestion with NcoI (New England Biolabs, Frankfurt/Main, Germany) was performed after ethanol precipitation of the PCR product. The precipitate was dissolved in 15-Al buffer (supplied by the manufacturer), 3 units enzyme were added and the mixture was incubated for 4 h at 37 jC. The last step was repeated after adding another 2 units of the enzyme. Electrophoresis of NcoI digestion of the PCR product of TNFa was performed on a gel containing 3.5% Metaphor agarose (Biozym, Hess. Oldendorf, Germany). The G allele produced two fragments with 20 and 87 bp, while the uncut A allele revealed a band corresponding to 107 bp. The NcoI digest of the LTa PCR product was separated by electrophoresis on 2% agarose. Two fragments of 545 and 196 bp were

22

S.G. Schwab et al. / Schizophrenia Research 65 (2003) 19–25

generated from the G allele. The uncut A allele revealed a band corresponding to 741 bp. 2.3. Statistical analysis The TDT (Spielman et al., 1993) was used to test for association between schizophrenia and TNFa or LTa in the sample of trio families. Exact two-sided P values for this test were calculated by comparing the observed number of transmissions of a specific allele to a binomial distribution with P = 0.5. An extension (Knapp, 2001) of the test of Zhao et al. (2000), which allows to test for association by using more than a single marker in general nuclear families, was used to assess the evidence for association between the disease and TNFa – LTa haplotypes in all three samples, and for the analysis of a single polymorphism in the sib pair sample as well as in the combined sample. This modified test results in a valid test of the null hypothesis ‘‘no association’’ even for samples consisting of families with more than one affected offspring. v2 tests were conducted in order to test the distribution of genotypes of pseudocontrols (obtained by combining the parental alleles not transmitted to the first affected offspring) for deviation from Hardy – Weinberg equilibrium. The genotype distribution in the combined sample was 80/77/22 ( P = 0.604) for the LTa polymorphism, and 8/58/132 ( P = 0.6116) for the TNFa polymorphism.

3. Results Due to failure in preparation or maintenance of permanent lymphoblast cell cultures, we were running out of genomic DNA for 13 families. Thus, the reduction of the genome scan sample of 71 families to 58 was unbiased in respect to the study. This sample was supplemented by 21 recently ascertained sib pair families from Southern Germany. Thus, a total number of 79 families were available for the sib pair sample. In addition to the sib pair sample, we used a sample of 128 trio families with schizophrenic offspring for internal confirmation. All trio families in this study had a family history of psychiatric disorders. We first analyzed the sib pair sample containing 79 families (Table 1). A biased transmission from heter-

Table 2 Transmission/disequilibrium test of genetic association (HA0 ) in presence of linkage between the TNFa and the LTa (TNFh) loci and schizophrenia ( P values) Polymorphism

Sib pairs

Trios

Combined

TNFa-G308A LTa (TNFh) + G252A TNFa + LTa

0.064 0.171 0.055

0.0225 0.114 0.03

0.003 0.031 0.001

HA0 Null Hypothesis no association.

ozygous parents (49 transmitted versus 31 not transmitted) was seen for the common G allele (TNF1) of the TNFa polymorphism (cut by the restriction enzyme NcoI) resulting in a P value of 0.064 (Table 2). In case of the LTa polymorphism, the common A allele (not cut by NcoI) was slightly more transmitted (56 transmitted versus 41 not transmitted, P = 0.171). These preliminary results were tested in the independently ascertained sample of 128 trio families. This sample produced biased transmission for the same alleles. The G allele of TNFa -G308A (45 transmitted versus 25 not transmitted, P = 0.022) and the A allele of LTa + G250A polymorphism (67 transmitted versus 49 not transmitted, P = 0.114) was preferentially transmitted from heterozygous parents (Table 2). Both samples combined revealed a P value of 0.003 for TNFa and a nominally significant P value of 0.03 for the LTa polymorphism (Table 2). Haplotypes, comprised of multiple markers, may increase the power to detect LD (Jorde, 2000; Zollner and von Haeseler, 2000). LD between the two polymorphisms was estimated in nontransmitted haplotypes using the classical DVstatistic (Lewontin 1988). A DVvalue of 0.77 ( P = 1.3e
07) was obtained for the sib pair sample and a DV value of 0.87 ( P = 1e
08) for the trio sample indicating a nearly complete LD as expected for two markers only approximately 2– 3 kb apart. TDT analysis using haplotypes resulted in a P value of 0.001 for the combined sample (Table 2).

4. Discussion The present study has two important features: (1) availability of two independent samples consisting of families with genetic history of psychiatric disorders provided the possibility of internal replication in a

S.G. Schwab et al. / Schizophrenia Research 65 (2003) 19–25

related sample; (2) using samples with parental genotypes enabled us to use the TDT. This family-based approach has been developed by Spielman et al. (1993) in order to avoid population association by stratification, which may confound case samples as well as control samples. Moreover, construction of haplotypes was facilitated in our sample by the availability of parental genotypes. Presence of stratification has been shown in empirical examples (Knowler et al., 1988; Reich et al., 1999). Although it has not been unambiguously demonstrated in disease association studies with case control design, it continues to be a matter of concern (Jorde, 2000; Cardon and Bell, 2001). In particular, polymorphisms with highly differing allele frequencies between populations may cause this effect. Ethnic diversity of the TNFa -G308A A allele frequency has been studied and found to be ranging from absent in a population of Pima Indians (Hamann et al., 1995) and in a Solomon population (Yoshida et al., 1998) to 1% in a Japanese and 7.5% in a Chinese population (Yoshida et al., 1998), while Caucasians range between 16% and 19% (Table 3). This demonstrates the importance of using a family-based approach with internal control as in the present study, thus avoiding population stratification. In this context, it should be noted that the Boin study (Boin et al., 2001) reports a

Table 3 Frequency of the A allele of the TNFa -G308A polymorphism in control populations

(Boin et al., 2001) (Wilson et al., 1992) (Walston et al., 1999) (Day et al., 1998) (Herrmann et al., 1998) (Rasmussen et al., 2000) Present papera Trios Sib pairsb

Population

n

Frequency of A allele

Northern Italian/ Caucasian not defined Caucasian African-American North-European Irish French Danish/Caucasian

138

0.110

40 362 62 126 97 376 380

0.160 0.170 0.180 0.182 0.242 0.157 0.189

Caucasian Caucasian

128 70

0.184 0.193

a Control alleles represent the nontransmitted alleles of the parents. b One randomly chosen offspring was considered for designing the transmitted alleles.

23

much lower frequency of allele A (11%) for the Caucasian/Italian control group than has been detected in other published Caucasian controls (Table 3). Whether the Mediterranean population has a lower frequency in general remains to be determined. We determined allele frequency in our samples from Southern Germany by counting the nontransmitted parental alleles, which can be considered as diseaseindependent control alleles (Falk and Rubinstein, 1987). Frequency of allele A was within the range of Caucasian control groups published in other studies (Table 3). Even though there is no support for existence of population-specific susceptibility genes in schizophrenia, we considered this possibility and eliminated the six families from Israel and the 10 families from Hungary before calculating TDT in the combined sample. A slightly less significant P value (0.004 for the haplotype) was obtained. This is apparently due to the reduction in sample size and does not point to different effects of the polymorphisms in German and non-German families. The possibility that the biased, but statistically not (in the sib pair sample) or only weak (in the trio sample) significant transmission that is observed when the two samples were calculated separately, may be due to chance, cannot be excluded. However, this becomes less likely, when the statistically significant values obtained in the combined sample ( P = 0.001 for the haplotype) are taken into consideration. There is evidence that allelic variants in the 5 Vuntranslated region of the proinflammatory cytokine TNFa are involved in susceptibility to a number of diseases, including multiple sclerosis, asthma, leprosy, leishmaniosis, and cerebral malaria. However, as shown by Knight et al. (1999), in the case of cerebral malaria, regulation of transcription is complex and may be influenced by a number of sequence variants in that region in an additive or interactive way. The role of TNFa in CNS is less clear. As discussed by Wassink et al. (2000) in the context of a study on heritability of the tumor necrosis factor receptor-II and its effect on brain morphology in schizophrenia, there is evidence for a neuroprotective role for TNFa (Cheng et al., 1994), but also for a neurotoxic effect on monoamine neurons in increased concentrations (Zhao and Schwartz, 1998).

24

S.G. Schwab et al. / Schizophrenia Research 65 (2003) 19–25

While these findings suggest the possibility that TNFa may play a role in schizophrenia, direct evidence for involvement in susceptibility to schizophrenia is relatively weak. Boin et al. (2001) found the rare, possibly transcription rate enhancing A allele associated with schizophrenia. In contrast to this study, we obtained evidence for association of the common G allele with schizophrenia in our sample of 207 families. If there is a functional association of the TNFa gene with schizophrenia based on this particular promoter polymorphism, it could be a different mechanism in our sample, perhaps due to a decrease rather than an increase in transcription rate. As shown by Beattie et al. (2002), a possible decrease in TNFa production may affect synaptic plasticity. TNFa may cause concentration-dependent changes in synaptic strength that occur during various forms of synaptic plasticity such as long-term potentiation and longterm depression (Beattie et al., 2002). Lastly, we cannot exclude that the investigated polymorphism might be in linkage disequilibrium with a different susceptibility allele of the TNFa gene or with another gene in the vicinity of TNFa. This possibility should be considered in particular since the HLA region is known to exhibit extensive linkage disequilibrium across the whole region. Further studies are needed in order to clarify whether this finding represents detection of a susceptibility allele for schizophrenia or is due to linkage disequilibrium with another susceptibility allele in the HLA region. In conclusion, our family-based association study revealed statistically significant biased transmission of the common G allele of the TNFa -G308A polymorphism in 207 families with schizophrenia combined from two independent samples with borderline significance. However, given the borderline significance level in the two independent samples, this finding remains preliminary and requires replication in other independent samples to exclude the possibility of representing a finding by chance.

Acknowledgements We wish to thank all patients and their family members, without their cooperation this work would not have been possible. This work was supported by Deutsche Forschungsgemeinschaft Sonderforschungs-

bereich SFB400 (W.M., M.R., D.B.W.), the German – Israeli Foundation for Scientific Research (B.L., D.B.W.) and Nationales Genomforschungsnetz, Germany (D.B.W.).

References Beattie, E.C., Stellwagen, D., Morishita, W., Bresnahan, J.C., Ha, B.K., Von Zastrow, M., et al., 2002. Control of synaptic strength by glial TNF alpha. Science 295, 2282 – 2285. Boin, F., Zanardini, R., Pioli, R., Altamura, C.A., Maes, M., Gennarelli, M., 2001. Association between -G308A tumor necrosis factor alpha gene polymorphism and schizophrenia. Mol. Psychiatry 6, 79 – 82. Brinkman, B.M., Zuijdeest, D., Kaijzel, E.L., Breedveld, F.C., Verweij, C.L., 1995. Relevance of the tumor necrosis factor alpha (TNF alpha) -308 promoter polymorphism in TNF alpha gene regulation. J. Inflam. 46, 32 – 41. Cardon, L.R., Bell, J.I., 2001. Association study designs for complex diseases. Nat. Rev., Genet. 2, 91 – 99. Cheng, B., Christakos, S., Mattson, M.P., 1994. Tumor necrosis factors protect neurons against metabolic – excitotoxic insults and promote maintenance of calcium homeostasis. Neuron 12, 139 – 153. Day, C.P., Grove, J., Daly, A.K., Stewart, M.W., Avery, P.J., Walker, M., 1998. Tumour necrosis factor-alpha gene promoter polymorphism and decreased insulin resistance. Diabetologia 41, 430 – 434. Falk, C.T., Rubinstein, P., 1987. Haplotype relative risks: an easy reliable way to construct a proper control sample for risk calculations. Ann. Hum. Genet. 51, 227 – 233. Fyer, A., Endicott, J., Mannuzza, S., Klein, D., 1985. Schedule for Affective Disorders and Schizophrenia-Lifetime Version (SADSLA). New York State Psychiatric Institute, New York. Ganguli, R., Yang, Z., Shurin, G., Chengappa, K.N., Brar, J.S., Gubbi, A.V., et al., 1994. Serum interleukin-6 concentration in schizophrenia: elevation associated with duration of illness. Psychiatry Res. 51, 1 – 10. Hamann, A., Mantzoros, C., Vidal-Puig, A., Flier, J.S., 1995. Genetic variability in the TNF-alpha promoter is not associated with type II diabetes mellitus (NIDDM). Biochem. Biophys. Res. Commun. 211, 833 – 839. Herrmann, S.M., Ricard, S., Nicaud, V., Mallet, C., Arveiler, D., Evans, A., et al., 1998. Polymorphisms of the tumour necrosis factor-alpha gene, coronary heart disease and obesity. Eur. J. Clin. Investig. 28, 59 – 66. Jorde, L.B., 2000. Linkage disequilibrium and the search for complex disease genes. Genome Res. 10, 1435 – 1444. Katila, H., Hanninen, K., Hurme, M., 1999. Polymorphisms of the interleukin-1 gene complex in schizophrenia. Mol. Psychiatry 4, 179 – 181. Knapp, M., 2001. A family-based test for association in the presence of linkage with multiple tightly linked markers in nuclear families with multiple affected children. Genet. Epidemiol. 21, 162.

S.G. Schwab et al. / Schizophrenia Research 65 (2003) 19–25 Knight, J.C., Udalova, I., Hill, A.V., Greenwood, B.M., Peshu, N., Marsh, K., et al., 1999. A polymorphism that affects OCT-1 binding to the TNF promoter region is associated with severe malaria. Nat. Genet. 22, 145 – 150. Knowler, W.C., Williams, R.C., Pettitt, D.J., Steinberg, A.G., 1988. Gm3;5,13,14 and type 2 diabetes mellitus: an association in American Indians with genetic admixture. Am. J. Hum. Genet. 43, 520 – 526. Lewontin, R.C., 1988. On measures of gametic disequilibrium. Genetics 120, 849 – 852. Licinio, J., Wong, M.L., 1997. Pathways and mechanisms for cytokine signaling of the central nervous system. J. Clin. Invest. 100, 2941 – 2947. Licinio, J., Seibyl, J.P., Altemus, M., Charney, D.S., Krystal, J.H., 1993. Elevated, C.S.F. levels of interleukin-2 in neuroleptic-free schizophrenic patients. Am. J. Psychiatry 150, 1408 – 1410. Lin, A., Kenis, G., Bignotti, S., Tura, G.J., De Jong, R., Bosmans, E., et al., 1998. The inflammatory response system in treatmentresistant schizophrenia: increased serum interleukin-6. Schizophr. Res. 32, 9 – 15. Maes, M., Bosmans, E., Calabrese, J., Smith, R., Meltzer, H.Y., 1995. Interleukin-2 and interleukin-6 in schizophrenia and mania: effects of neuroleptics and mood stabilizers. J. Psychiatr. Res. 29, 141 – 152. Maes, M., Bocchio Chiavetto, L., Bignotti, S., Battisa Tura, G., Pioli, R., Boin, F., et al., 2000. Effects of atypical antipsychotics on the inflammatory response system in schizophrenic patients resistant to treatment with typical neuroleptics. Eur. Neuropsychopharmacol. 10, 119 – 124. McGuffin, P., Farmer, A., Harvey, I., 1991. A polydiagnostic application of operational criteria in studies of psychotic illness. Development and reliability of the OPCRIT system. Arch. Gen. Psychiatry 48, 764 – 770. Messer, G., Spengler, U., Jung, M.C., Honold, G., Blomer, K., Pape, G.R., et al., 1991. Polymorphic structure of the tumor necrosis factor (TNF) locus: an NcoI polymorphism in the first intron of the human TNF-beta gene correlates with a variant amino acid in position 26 and a reduced level of TNF-beta production. J. Exp. Med. 173, 209 – 219. Moffatt, M.F., Cookson, W.O., 1997. Tumour necrosis factor haplotypes and asthma. Hum. Mol. Genet. 6, 551 – 554. Monteleone, P., Fabrazzo, M., Tortorella, A., Maj, M., 1997. Plasma levels of interleukin-6 and tumor necrosis factor alpha in chronic schizophrenia: effects of clozapine treatment. Psychiatry Res. 71, 11 – 17. Naudin, J., Capo, C., Giusano, B., Mege, J.L., Azorin, J.M., 1997. A differential role for interleukin-6 and tumor necrosis factoralpha in schizophrenia? Schizophr. Res. 26, 227 – 233. Pociot, F., Briant, L., Jongeneel, C.V., Molvig, J., Worsaae, H., Abbal, M., et al., 1993. Association of tumor necrosis factor (TNF) and class II major histocompatibility complex alleles with the secretion of TNF-alpha and TNF-beta by human mononuclear cells: a possible link to insulin-dependent diabetes mellitus. Eur. J. Immunol. 23, 224 – 231. Rasmussen, S.K., Urhammer, S.A., Jensen, J.N., Hansen, T., BorchJohnsen, K., Pedersen, O., 2000. The -238 and -308 G ! A polymorphisms of the tumor necrosis factor alpha gene pro-

25

moter are not associated with features of the insulin resistance syndrome or altered birth weight in Danish Caucasians. J. Clin. Endocrinol. Metab. 85, 1731 – 1734. Reich, T., Hinrichs, A., Culverhouse, R., Bierut, L., 1999. Genetic studies of alcoholism and substance dependence. Am. J. Hum. Genet. 65, 599 – 605. Schwab, S.G., Hallmayer, J., Albus, M., Lerer, B., Eckstein, G.N., Borrmann, M., et al., 2000. A genome-wide autosomal screen for schizophrenia susceptibility loci in 71 families with affected siblings: support for loci on chromosome 10p and 6. Mol. Psychiatry 5, 638 – 649. Schwab, S.G., Hallmayer, J., Freimann, J., Lerer, B., Albus, M., Borrmann-Hassenbach, M., et al., 2002. Investigation of linkage and association/linkage disequilibrium of HLA A-, DQA1-, DQB1-, and DRB1- alleles in 69 sib-pair- and 89 trio-families with schizophrenia. Am. J. Med. Genet. 114, 315 – 320. Spielman, R.S., McGinnis, R.E., Ewens, W.J., 1993. Transmission test for linkage disequilibrium: the insulin gene region and insulin-dependent diabetes mellitus (IDDM). Am. J. Hum. Genet. 52, 506 – 516. Spitzer, R., Endicott, J., Robins, E., 1978. Research Diagnostic Criteria for a Selected Group of Functional Disorders. New York State Psychiatric Institute, New York. Walston, J., Seibert, M., Yen, C.J., Cheskin, L.J., Andersen, R.E., 1999. Tumor necrosis factor-alpha-238 and -308 polymorphisms do not associated with traits related to obesity and insulin resistance. Diabetes 48, 2096 – 2098. Wassink, T.H., Crowe, R.R., Andreasen, N.C., 2000. Tumor necrosis factor receptor-II: heritability and effect on brain morphology in schizophrenia. Mol. Psychiatry 5, 678 – 682. Wilson, A.G., di Giovine, F.S., Blakemore, A.I., Duff, G.W., 1992. Single base polymorphism in the human tumour necrosis factor alpha (TNF alpha) gene detectable by NcoI restriction of PCR product. Hum. Mol. Genet. 1, 353. Wilson, A.G., Symons, J.A., McDowell, T.L., McDevitt, H.O., Duff, G.W., 1997. Effects of a polymorphism in the human tumor necrosis factor alpha promoter on transcriptional activation. Proc. Natl. Acad. Sci. U. S. A. 94, 3195 – 3199. Wong, M.-L., Licinio, J., 1994. Localization of stem cell factor mRNA in adult rat hippocampus. Neuroimmunomodulation 1, 181 – 187. Wright, P., Nimgaonkar, V.L., Donaldson, P.T., Murray, R.M., 2001. Schizophrenia and HLA: a review. Schizophr. Res. 47, 1 – 12. Yoshida, A., Ishii, A., Leafasia, J.L., Zhou, D., Chen, Y., Tang, L., et al., 1998. Polymorphism of the tumor necrosis factor A gene in Melanesian and East-Asian populations. Immunogenetics 47, 497 – 498. Zhao, B., Schwartz, J.P., 1998. Involvement of cytokines in normal CNS development and neurological diseases: recent progress and perspectives. J. Neurosci. Res. 52, 7 – 16. Zhao, H., Zhang, S., Merikangas, K.R., Trixler, M., Wildenauer, D.B., Sun, F., et al., 2000. Transmission/disequilibrium tests using multiple tightly linked markers. Am. J. Hum. Genet. 67, 936 – 946. Zollner, S., von Haeseler, A., 2000. A coalescent approach to study linkage disequilibrium between single-nucleotide polymorphisms. Am. J. Hum. Genet. 66, 615 – 628.