A Single Nucleotide Insertion in Codon 317 of the CD36 Gene Leads to CD36 Deficiency
CD36 is a multifunctional integral-membrane glycoprotein that acts as a receptor for thrombospondin, collagen, long-chain fatty acids, and oxidized LDL. Platelet CD36 deficiency can be divided into two groups. In type I, neither platelets nor monocytes/macrophages express CD36; in type II, monocytes/macrophages express CD36 but platelets do not. Two known mutations cause CD36 deficiency, ie, a 478C→T substitution in codon 90 (proline90→serine) and a dinucleotide deletion at nucleotide 539 in codon 110. In this study we investigated a type I Japanese subject (A.T.) and identified a new mutation, a single nucleotide insertion at nucleotide 1159 in codon 317. This mutation leads to a frameshift and the appearance of a premature stop codon. CD36 gene analysis indicated that A.T. was a compound heterozygote for a dinucleotide deletion at nucleotide 539 and the single nucleotide insertion at nucleotide 1159. RNase protection studies suggested that the new mutation as well as the dinucleotide deletion led to a marked reduction in the level of CD36 mRNA in her macrophages. However, the new mutation could be detected in macrophage but not platelet CD36 mRNA. These data suggest that the allele having the single nucleotide insertion in this subject has an additional abnormality that results in the absence of the mutated CD36 mRNA in platelets.
- Received August 16, 1995.
- Revision received April 9, 1996.
The 88-kD integral-membrane glycoprotein CD36 is expressed on monocytes/macrophages, platelets, erythroblasts, capillary endothelial cells, and mammary epithelial cells.1 CD36 has been proposed as an adhesive receptor for collagen and thrombospondin and as a transporter for long-chain fatty acids. CD36 also mediates cytoadherence of Plasmodium falciparum–infected erythrocytes to the endothelium.2 3 4 5 Furthermore, CD36 functions as a scavenger receptor on macrophages for oxidized LDL.6 7
Human CD36 cDNA has been isolated from placenta, platelets, and HEL cells.8 9 10 11 The CD36 gene is located on chromosome 7q11.12,11 12 and the structural organization of the human CD36 gene, which is composed of 12 coding exons that cover >32 kb of genomic DNA, has been reported.11 13 14
CD36 deficiency was first identified in a patient who showed refractoriness to HLA-matched platelet transfusions.15 16 17 We18 and others19 have demonstrated that CD36 deficiency can be divided into two subgroups. In type I, neither platelets nor monocytes/macrophages express CD36; in type II, monocytes/macrophages express CD36 but platelets do not. Type II subjects are identified in ≈3% of blood donors in the Japanese population15 and ≈0.3% of the US population.17 The incidence of type I subjects has not been determined. However, our preliminary study suggests that the incidence of type I deficiency may be ≤10% that of type II deficiency. The individuals affected are apparently healthy and suffer no hemostatic problems in spite of the absence of platelet CD36. However, we recently demonstrated a marked reduction in the uptake of oxidized LDL by CD36-deficient macrophages, which suggests that there may be some differences in atherogenesis between type I CD36-deficient subjects and CD36-positive subjects.20
We have reported two mutations that are responsible for CD36 deficiency, ie, a substitution of T for C at nt 478 of CD36 cDNA in codon 90 (proline90→serine)21 22 and a dinucleotide deletion at nt 539 in codon 110.23 In addition to these two mutations, comparison of platelet and monocyte/macrophage CD36 mRNA suggests that an allele having a platelet-specific mRNA transcription defect(s) may be involved in type II subjects.24
In this study we report a new mutation, a single nucleotide insertion at nt 1159 in codon 317, which results in a CD36 deficiency in a type I Japanese subject. This mutation causes a frameshift and leads to the appearance of a premature stop codon. RNase protection studies suggest that the new mutation results in a marked reduction in the level of macrophage CD36 mRNA. However, we could detect the mutation in macrophage but not platelet CD36 mRNA. These data suggest that the allele with the insertion in this subject has an additional abnormality that leads to the absence of the mutated CD36 mRNA in platelets.
A type I CD36-deficient subject (A.T.) was found by screening with anti-CD36 (anti-Naka) alloantibodies in the Osaka Red Cross Blood Center. By using flow cytometric and immunohistochemical analyses, we have already demonstrated that neither the platelets nor the monocyte-derived macrophages from this individual express CD36 on their surface.20 Platelets,25 mononuclear cells,25 and monocyte-derived macrophages20 were isolated. In brief, mononuclear cells were plated in RPMI 1640 containing 10% (vol/vol) human type AB serum. After 4 hours of incubation at 37°C in 5% CO2, nonadherent cells were removed. After a 14-day culture, macrophages were collected by using a cell scraper, and total cellular RNA and DNA were obtained.21 22 23 24 25
Amplification of CD36 cDNA and Nucleotide Sequencing
PCR amplification of platelet and macrophage CD36 cDNA was performed as described21 22 23 24 25 with some modifications. In brief, first-strand cDNA was synthesized from total cellular RNA by using Moloney murine leukemia virus reverse transcriptase (Bethesda Research Laboratories) and 100 pmol random hexamer oligonucleotides (Takara Shuzo Corp). First-round PCR was performed by using primers NAK5 (sense, nt 8 through 32) and NAK2 (antisense, nt 1045 through 1025) for the 5′ region or NAK3 (sense, nt 932 through 953) and NAK4 (antisense, nt 1745 through 1726) for the 3′ region. Second-round PCR was performed by using first-round PCR products as a template with nested primers: NAK18-HIND (sense, nt 18 through 37) and NAK2′ (antisense, nt 1009 through 982) for the 5′ region or NAK3′ (sense, nt 962 through 977) and NAK4′ (antisense, nt 1714 through 1698) for the 3′ region. The nucleotide sequence of primers and PCR conditions have been described.21 22 23 24 25 PCR products were subcloned into pBluescriptII (Stratagene), and nucleotide sequence analysis was performed by using the Sequenase kit, version 2 (US Biochemical Corp).
Restriction Enzyme Digestion of the Amplified Products
Amplification around the deletion in CD36 cDNA was performed by using the first-round PCR products (nt 8 through 1045) as a template with nested primers: NAK7 (sense, 293CAGTTGGAGACCTGCTTATC) and SSPI(−) (antisense, nt 567 through 542). Amplification around the dinucleotide deletion of the CD36 gene was performed by using primers ID(+) and SSPI(−).23
Amplification of CD36 cDNA around the single nucleotide insertion in codon 317 was performed by using the first-round PCR products (nt 932 through 1745) as a template with nested primers: XMN(+) (sense, 1056CGTTAATCTGAAAGGATTCC) and XMN(−) (antisense, 1179TGTACAATTTTTTGAGAGAA). The nucleotides underlined are mismatched. Amplification of the CD36 gene around the insertion was performed by using ≈1 μg DNA as a template with the primers XMN(+) and XMN(−). The amplification was performed for 30 cycles in a DNA thermal cycler (GeneAmp PCR System 9600, Perkin-Elmer Co) at 94°C for 0.5 minute, 50°C for 2 minutes, and 72°C for 1 minute.
The amplified fragments were directly digested with Ssp I (Toyobo) and Xmn I (New England BioLabs) for the detection of the dinucleotide deletion and the single nucleotide insertion, respectively. The resulting fragments were subjected to electrophoresis on nondenatured polyacrylamide gels and stained with ethidium bromide.
Detection of TG Repeat Polymorphism of the CD36 Gene
Quantification of mRNA by Using RT-PCR or RNase Protection Assay
Semiquantification of CD36 mRNA was performed by using RT-PCR23 27 with some modifications. In brief, first-strand cDNA was synthesized from 250 ng total cellular RNA with random hexamer oligonucleotides. Subsequently, primers and Taq polymerase were added to the reaction mixture, and the mixture was divided in aliquots and placed in four separate tubes. Primers NAK5 and NAK6 (antisense, nt 270 through 250) were used for CD36 cDNA, and C-1 (sense, nt 103 through 122) and C-2 (antisense, nt 642 through 619) for β-actin cDNA. Amplification was performed at 94°C for 0.5 minute, 50°C for 1 minute, and 72°C for 1 minute. Reaction tubes were removed at 18, 21, 24, and 27 cycles, and the amplified products were subjected to electrophoresis on 1.5% agarose gels and stained with ethidium bromide.
Quantification of CD36 mRNA was further performed by using an RNase protection assay.28 CD36 cDNA corresponding to nt 517 through 918 was cloned into pGEM-T vector (Promega Corp) and linearized by using EcoRI. An antisense RNA was transcribed by using T7 RNA polymerase in the presence of [32P]UTP (3000 Ci/mmol, New England Nuclear) and used as a probe for the RNase protection assay. Radiolabeled antisense RNA for human β-actin was used as an internal standard. Total RNA (10 μg) from macrophages of A.T. or a CD36-positive subject was hybridized with 4×104 cpm of each probe at 42°C overnight. Annealing products were digested with RNase T1 at 37°C for 30 minutes. The protected fragments were precipitated and subjected to 6% polyacrylamide/urea gel. The results were analyzed by using autoradiography and quantified by laser densitometric scanning.
Detection of the Dinucleotide Deletion in CD36 cDNA
Amplification of the 5′ region of platelet CD36 mRNA from A.T. showed an apparently normal-sized band and a smaller band of ≈150 bp (Fig 1⇓). In contrast, amplification of the 5′ region of macrophage CD36 mRNA showed only normal-sized CD36 cDNA. Amplification of the 3′ region of CD36 mRNA showed only normal-sized CD36 cDNA in both platelets and macrophages. We have already demonstrated a dinucleotide deletion at nt 539 in a type I CD36-deficient subject, Y.K.23 Interestingly, platelets and macrophages from Y.K. contained two different-sized CD36 mRNAs, one with the dinucleotide deletion and the other with the dinucleotide deletion and a 161-bp deletion corresponding to a loss of exon IV.23 Since the sizes of PCR products from the 5′ region of platelet CD36 mRNA from A.T. were very similar to those from Y.K., we investigated whether the dinucleotide deletion might exist in A.T.'s platelet CD36 mRNA. To detect the deletion, we designed a primer, SSPI(−), so that only fragments from nt 293 through 567 having the dinucleotide deletion could be digested with Ssp I (Fig 2A⇓).23 In A.T.'s platelets, both the apparently normal-sized and ≈150-bp smaller-sized cDNAs were again amplified when the primers NAK7 and SSPI(−) were used. These amplified CD36 cDNA fragments from A.T.'s platelet mRNA were completely digested with Ssp I, indicating that all her platelet CD36 mRNA has the dinucleotide deletion (Fig 2B⇓). These results also suggested that the ≈150-bp deletion existed between nt 293 and 539, which would probably correspond to the 161-bp deletion (nt 331 through 491) observed in Y.K.23 In contrast, amplified CD36 cDNA from A.T.'s macrophage mRNA was only partially digested with Ssp I, suggesting that there were two types of CD36 mRNAs in her macrophages: one with the dinucleotide deletion and the other without.
Nucleotide Sequence of Macrophage CD36 cDNA
To identify the abnormality in A.T.'s macrophage CD36 mRNA without the dinucleotide deletion, we subcloned and sequenced her macrophage CD36 cDNA. In two of three clones a single nucleotide, A, was inserted at nt 1159 (Fig 3⇓). This insertion produced a frameshift that resulted in a premature termination at nt 1263. To confirm the insertion, we amplified CD36 cDNA around nt 1159 by using the primer XMN(−), which was designed to create an Xmn I site at nt 1157. If the single nucleotide were inserted, then the Xmn I site (GAANN↓NNTTC) would be destroyed (Fig 4A⇓). Platelet CD36 cDNA from this subject was completely digested with Xmn I, indicating that platelet CD36 mRNA did not have the insertion (Fig 4B⇓). These data are consistent with those obtained from the Ssp I digestion. In A.T.'s macrophages, however, Xmn I digestion showed a heterozygous pattern, indicating that both CD36 mRNA with and without the insertion existed in her macrophages.
Analysis of the CD36 Gene
Ssp I digestion of the amplified CD36 gene fragments around exon V showed that A.T. was heterozygous for the dinucleotide deletion (Fig 5A⇓). The single nucleotide insertion is supposed to exist in exon X of the CD36 gene.14 Xmn I digestion of the CD36 gene fragments around the insertion also showed that she was heterozygous for the insertion (Fig 5B⇓). Both her husband and daughter were CD36-positive subjects. DNA analysis showed that the CD36 gene of the daughter had the insertion but not the dinucleotide deletion. Neither the insertion nor the deletion existed in the husband's CD36 gene (data not shown). These data indicate that the insertion and the deletion exist in different alleles and that A.T. is a compound heterozygote for the single nucleotide insertion and the dinucleotide deletion. Her daughter is a carrier of the single nucleotide insertion. We also examined a TG repeat polymorphism in intron C of the CD36 gene in A.T.'s family members. Genotypes of A.T., her husband, and daughter were A3/A5, A3/A3, and A3/A5, respectively (data not shown). These data suggest that the single nucleotide insertion was linked to the A5 allele.
Quantification of Macrophage CD36 mRNA
We have demonstrated that the dinucleotide deletion leads to a decrease in CD36 mRNA levels,23 and nonsense or frameshift mutations often affect mRNA expression levels.29 Therefore, we investigated whether the single nucleotide insertion might affect CD36 mRNA levels in macrophages. First, we performed an RT-PCR assay on total RNA obtained from macrophages of A.T.'s daughter, who is a carrier of the insertion. We were unable to amplify CD36 cDNA having the single nucleotide insertion from the macrophage mRNA, suggesting that the level of the CD36 mRNA with the single nucleotide insertion was too small to be amplified in the presence of normal CD36 mRNA (data not shown). In A.T.'s macrophages, we assumed that the decrease in the level of the transcripts from the allele with the dinucleotide deletion enabled us to amplify CD36 mRNA with the insertion.23
To confirm our assumption, we performed semiquantification of CD36 mRNA by using an RT-PCR assay with β-actin mRNA as an internal control. The intensity of β-actin– and CD36 cDNA–specific bands increased as cycle numbers increased in CD36-positive macrophages (Fig 6A⇓). In A.T.'s macrophages the amplification of CD36 cDNA bands was retarded compared with that of β-actin cDNA bands, indicating that CD36 mRNA levels were reduced in A.T.'s macrophages. RNase protection studies further confirmed that the CD36 mRNA levels in A.T.'s macrophages were markedly reduced and that the amount was ≈10% of CD36-positive macrophages (Fig 6B⇓). These data demonstrate that the single nucleotide insertion as well as the dinucleotide deletion may lead to marked reductions in the level of CD36 mRNA.
CD36 has been implicated as a receptor for oxidized LDL.6 7 By employing CD36-deficient macrophages from two type I CD36-deficient subjects, including A.T., we have clearly shown that CD36 can account for up to 40% of the binding and internalization of oxidized LDL by macrophages.20 In addition, we have shown that cholesteryl ester mass accumulation in macrophages after incubation with oxidized LDL is reduced by ≈40% due to CD36 deficiency.20 Thus, CD36 may be one of the physiological receptors for oxidized LDL and may contribute in part to foam cell formation in atherosclerotic lesions.
In this study we demonstrated a new mutation in the CD36 gene, a single nucleotide insertion at nt 1159 in codon 317, which leads to CD36 deficiency and produces a frameshift that results in a premature termination at nt 1263. A type I CD36-deficient subject, A.T., is a compound heterozygote for the single nucleotide insertion and the previously described dinucleotide deletion at nt 539 in codon 110, and we confirmed the hereditary nature of the insertion by the analysis of A.T.'s family members. The amount of A.T.'s CD36 mRNA was ≈10% that of a CD36-positive control subject. These data suggest that the insertion as well as the dinucleotide deletion leads to marked reductions in CD36 mRNA levels in monocytes/macrophages. Interestingly, the CD36 mRNA with the single nucleotide insertion was detected in A.T.'s macrophages but not her platelets, suggesting that the allele with the insertion has an additional abnormality that results in the absence of the mutated CD36 mRNA in platelets.
There are some discrepancies between the results of Ssp I digestion and Xmn I digestion of the amplified CD36 genomic DNA (and cDNA) from A.T.'s macrophages. The Ssp I digestion showed that the majority of CD36 mRNA in her macrophages may not have the dinucleotide deletion, whereas the Xmn I digestion did not show such an apparent difference between these mutated CD36 mRNA levels (compare Figs 2B and 4B⇑⇑). In addition, the Ssp I digestion still showed the apparent difference between the amount of genomic DNA with and without the deletion, which conflicts with the fact that A.T. is heterozygous for the dinucleotide deletion (Fig 5⇑). These data indicate that there are some differences between the amplification efficiency of wild-type and mutated cDNA (and DNA) in the case of the primer SSPI(−), the mismatched primer just adjacent to the dinucleotide deletion. From the data obtained from the Xmn I digestion, we think that the level of the CD36 mRNA with the insertion may be similar to that with the deletion in A.T.'s macrophages.
Transcripts that result in an early termination due to nonsense or frameshift mutations often lead to marked reductions in the steady-state level of cytoplasmic mRNA.29 30 Therefore, it is likely that the single nucleotide insertion as well as the dinucleotide deletion causes the reduction in the level of CD36 mRNA in A.T.'s macrophages. However, the molecular mechanism by which the different CD36 mRNA expression occurs in A.T.'s platelets and macrophages is unclear. One possible explanation is that the CD36 mRNA with the insertion may be unstable in both platelets and macrophages so that the steady-state mRNA level is very low.30 Under these conditions, the mutated CD36 mRNA could be more unstable in platelets than in macrophages, since platelets circulate for 10 days with no new mRNA being made. However, the presence of CD36 mRNA with the dinucleotide deletion both in platelets and macrophages could not be explained by this hypothesis. An alternative explanation is that the insertion may be linked to another mutation that affects transcriptional efficiency in platelets. We have already revealed the different expression of CD36 mRNA between platelets and monocytes/macrophages from some type II CD36-deficient subjects who are heterozygous for the mutation (Ser90) in the CD36 gene.22 24 However, only the mutated type (Ser90) of CD36 mRNA was detected in their platelets, whereas both wild-type (Pro90) and the mutated type were detected in monocytes/macrophages. From these data, we hypothesized that in these subjects the allele without the Ser90 mutation had a platelet-specific transcriptional defect(s) (“platelet-specific silent” allele).24 The platelet-specific silent allele appears widely distributed in the Japanese population, since type II subjects are identified in ≈3% of Japanese blood donors.15 Therefore, it is possible that the single nucleotide insertion has occurred in the platelet-specific silent allele in A.T. We have suggested that the platelet-specific silent allele is linked to a specific allele (A5 allele) of a TG repeat polymorphism in intron C of the CD36 gene.24 26 DNA analysis of A.T.'s family members suggested that the insertion was present in the A5 allele in A.T. These results support our hypothesis that the insertion has occurred in the platelet-specific silent allele.
Pearce et al31 have shown that CD36 has only one transmembrane domain at the C-terminal end by deletion mutagenesis and that a functional truncated CD36 lacking the C-terminal transmembrane domain is secreted into the postculture medium by transfected cells. Since the single nucleotide insertion causes a frameshift relatively near the C-terminal transmembrane domain, a truncated form of CD36 could be secreted in plasma or tissue. In A.T., however, the amount of the truncated form is supposed to be very low, because the allele having the insertion is transcribed only in monocytes/macrophages, and the CD36 mRNA level was markedly reduced. Therefore, even if the truncated CD36 could exist, it is unlikely that the secreted form of CD36 could have a physiological or pathological role in this subject.
In summary, we have shown that the single nucleotide insertion in codon 317 of the CD36 gene is associated with CD36 deficiency. The allele having this mutation is also likely to have a platelet-specific transcriptional defect(s).
Selected Abbreviations and Acronyms
|HEL||=||human erythroid leukemia cell line|
|PCR||=||polymerase chain reaction|
|RT-PCR||=||reverse transcription polymerase chain reaction|
This work was supported in part by grants-in-aid for science research from the Ministry of Education, Science, and Culture in Japan. We thank Dr Robert H. Lipsky, American Red Cross, NY, for giving us information about the CD36 gene organization before its publication, and Dr Sanford J. Shattil, The Scripps Research Institute, La Jolla, Calif, for his helpful suggestions. We also thank Drs Itaru Matsumura and Kouji Miyaoka for their valuable assistance.
Greenwalt DE, Lipsky RH, Ockenhouse CF, Ikeda H, Tandon NN, Jamieson GA. Membrane glycoprotein CD36: a review of its roles in adherence, signal transduction, and transfusion medicine. Blood. 1992;80:1105-1115.
Tandon NN, Kralisz U, Jamieson GA. Identification of GP IV (CD36) as a primary receptor for platelet-collagen adhesion. J Biol Chem. 1989;264:7576-7583.
Asch AS, Barnwell J, Silverstein RL, Nachman RL. Isolation of the thrombospondin membrane receptor. J Clin Invest. 1987;79:1054-1061.
Abumrad NA, El-Maghrabi MR, Amri EZ, Lopez E, Grimaidi PA. Cloning of a rat adipocyte membrane protein implicated in binding or transport of long-chain fatty acids that is induced during preadipocyte differentiation. J Biol Chem. 1993;268:17665-17668.
Ockenhouse CF, Tandon NN, Magowan C, Jamieson GA, Chulay J. Identification of a platelet membrane glycoprotein as a falciparum malaria sequestration receptor. Science. 1989;243:1469-1471.
Endemann G, Stanton LW, Madden KS, Bryant CM, White RT, Protter AA. CD36 is a receptor for oxidized low density lipoprotein. J Biol Chem. 1993;268:11811-11816.
Nicholson AC, Frieda S, Pearce A, Silverstein RL. Oxidized LDL binds to CD36 on human monocyte–derived macrophages and transfected cell lines. Arterioscler Thromb Vasc Biol. 1995;15:269-275.
Tang Y, Taylor KT, Sobieski DA, Medved ES, Lipsky RH. Identification of a human CD36 isoform produced by exon skipping: conservation of exon organization and pre-mRNA splicing patterns with a CD36 gene family member, CLA-1. J Biol Chem. 1994;269:6011-6015.
Armesilla AL, Vega GA. Structural organization of the gene for human CD36 glycoprotein. J Biol Chem. 1994;269:18985-18991.
Tomiyama Y, Take H, Ikeda H, Mitani T, Furubayashi T, Mizutani H, Yamamoto N, Tandon NN, Sekiguchi S, Jamieson GA, Kurata T, Yonezawa T, Tarui S. Identification of the platelet-specific alloantigen, Naka, on platelet membrane glycoprotein IV. Blood. 1990;75:684-687.
Yamamoto N, Ikeda H, Tandon NN, Herman J, Tomiyama Y, Mitani T, Sekiguchi S, Lipsky R, Kralisz U, Jamieson GA. A platelet membrane glycoprotein (GP) deficiency in healthy blood donors: Naka- platelets lack detectable GP IV (CD36). Blood. 1990;76:1698-1703.
Yamamoto N, Akamatsu N, Sakuraba H, Yamazaki H, Tanoue K. Platelet glycoprotein IV (CD36) deficiency is associated with the absence (type I) or the presence (type II) of glycoprotein IV on monocytes. Blood. 1994;83:392-397.
Nozaki S, Kashiwagi H, Yamashita S, Nakagawa T, Kostner B, Tomiyama Y, Nakata A, Ishigami M, Miyagawa J, Kameda-Takemura K, Kurata Y, Matsuzawa Y. Reduced uptake of oxidized low density lipoproteins in monocyte-derived macrophages from CD36 deficient subjects. J Clin Invest. 1995;96:1859-1865.
Kashiwagi H, Tomiyama Y, Honda S, Kosugi S, Shiraga M, Nagao N, Sekiguchi S, Kanayama Y, Kurata Y, Matsuzawa Y. Molecular basis of CD36 deficiency: evidence that a 478C→T substitution (proline90→serine) in CD36 cDNA accounts for CD36 deficiency. J Clin Invest. 1995;95:1040-1046.
Kashiwagi H, Tomiyama Y, Kosugi S, Shiraga M, Lipsky RH, Kanayama Y, Kurata Y, Matsuzawa Y. Identification of molecular defects in a subject with type I CD36 deficiency. Blood. 1994;83:3545-3552.
Kashiwagi H, Tomiyama Y, Kosugi S, Shiraga M, Lipsky RH, Nagao N, Kanakura Y, Kurata Y, Matsuzawa Y. Family studies of type II CD36 deficient subjects: linkage of a CD36 allele to a platelet specific mRNA expression defect(s) causing type II CD36 deficiency. Thromb Haemost. 1995;74:758-763.
Lipsky RH, Ikeda H, Medved ES. A dinucleotide repeat in the third intron of CD36. Hum Mol Genet. 1994;3:217.
Inoue K, Sugiyama H, Ogawa H, Yamagami T, Azuma T, Oka Y, Miwa H, Kita K, Hiraoka A, Masaoka T, Nasu K, Kyo T, Dohy H, Hara J, Kanamaru A, Kishimoto T. Expression of the interleukin-6 (IL-6), IL-6 receptor, and gp130 genes in acute leukemia. Blood. 1994;84:2672-2680.
Funahashi T, Shimomura I, Hiraoka H, Arai T, Takahashi M, Nakamura T, Nozaki S, Yamashita S, Takemura K, Tokunaga K, Matsuzawa Y. Enhanced expression of rat obese (ob) gene in adipose tissues of ventromedial hypothalamus (VMH)-lesioned rats. Biochem Biophys Res Commun. 1995;211:469-475.
Pearce SFA, Wu J, Silverstein RL. A carboxy terminal truncation mutant of CD36 is secreted and binds thrombospondin: evidence for a single transmembrane domain. Blood. 1994;84:384-389.