Institute of Cytology and Genetics

An annotated reference sequence representing the hexaploid bread wheat genome in 21 pseudomolecules has been analyzed to identify the distribution and genomic context of coding and noncoding elements across the A, B, and D subgenomes. With an estimated coverage of 94% of the genome and containing 107,891 high-confidence gene models, this assembly enabled the discovery of tissue- and developmental stage-related coexpression networks by providing a transcriptome atlas representing major stages of wheat development. Dynamics of complex gene families involved in environmental adaptation and end-use quality were revealed at subgenome resolution and contextualized to known agronomic single-gene or quantitative trait loci. This community resource establishes the foundation for accelerating wheat research and application through improved understanding of wheat biology and genomics-assisted breeding.

Anopheles gambiae is the principal vector of malaria, a disease that afflicts more than 500 million people and causes more than 1 million deaths each year. Tenfold shotgun sequence coverage was obtained from the PEST strain of A. gambiae and assembled into scaffolds that span 278 million base pairs. A total of 91% of the genome was organized in 303 scaffolds; the largest scaffold was 23.1 million base pairs. There was substantial genetic variation within this strain, and the apparent existence of two haplotypes of approximately equal frequency ("dual haplotypes") in a substantial fraction of the genome likely reflects the outbred nature of the PEST strain. The sequence produced a conservative inference of more than 400,000 single-nucleotide polymorphisms that showed a markedly bimodal density distribution. Analysis of the genome sequence revealed strong evidence for about 14,000 protein-encoding transcripts. Prominent expansions in specific families of proteins likely involved in cell adhesion and immunity were noted. An expressed sequence tag analysis of genes regulated by blood feeding provided insights into the physiological adaptations of a hematophagous insect.

Deep sequencing of the gut microbiomes of 1135 participants from a Dutch population-based cohort shows relations between the microbiome and 126 exogenous and intrinsic host factors, including 31 intrinsic factors, 12 diseases, 19 drug groups, 4 smoking categories, and 60 dietary factors. These factors collectively explain 18.7% of the variation seen in the interindividual distance of microbial composition. We could associate 110 factors to 125 species and observed that fecal chromogranin A (CgA), a protein secreted by enteroendocrine cells, was exclusively associated with 61 microbial species whose abundance collectively accounted for 53% of microbial composition. Low CgA concentrations were seen in individuals with a more diverse microbiome. These results are an important step toward a better understanding of environment-diet-microbe-host interactions.

Here we report the Simons Genome Diversity Project data set: high quality genomes from 300 individuals from 142 diverse populations. These genomes include at least 5.8 million base pairs that are not present in the human reference genome. Our analysis reveals key features of the landscape of human genome variation, including that the rate of accumulation of mutations has accelerated by about 5% in non-Africans compared to Africans since divergence. We show that the ancestors of some pairs of present-day human populations were substantially separated by 100,000 years ago, well before the archaeologically attested onset of behavioural modernity. We also demonstrate that indigenous Australians, New Guineans and Andamanese do not derive substantial ancestry from an early dispersal of modern humans; instead, their modern human ancestry is consistent with coming from the same source as that of other non-Africans. Deep whole-genome sequencing of 300 individuals from 142 diverse populations provides insights into key population genetic parameters, shows that all modern human ancestry outside of Africa including in Australasians is consistent with descending from a single founding population, and suggests a higher rate of accumulation of mutations in non-Africans compared to Africans since divergence. Three international collaborations reporting in this issue of Nature describe 787 high-quality genomes from individuals from geographically diverse populations. David Reich and colleagues analysed whole-genome sequences of 300 individuals from 142 populations. Their findings include an accelerated estimated rate of accumulation of mutations in non-Africans compared to Africans since divergence, and that indigenous Australians, New Guineans and Andamanese do not derive substantial ancestry from an early dispersal of modern humans but from the same source as that of other non-Africans. Eske Willerlsev and colleagues obtained whole-genome data for 83 Aboriginal Australians and 25 Papuans from the New Guinea Highlands. They estimate that Aboriginal Australians and Papuans diverged from Eurasian populations 51,000–72,000 years ago, following a single out-of-Africa dispersal. Luca Pagani et al. report on a dataset of 483 high-coverage human genomes from 148 populations worldwide, including 379 new genomes from 125 populations. Their analyses support the model by which all non-African populations derive most of their genetic ancestry from a single recent migration out of Africa, although a Papuan contribution suggests a trace of an earlier human expansion.

We report here the genome sequence of an ancient human. Obtained from ∼4,000-year-old permafrost-preserved hair, the genome represents a male individual from the first known culture to settle in Greenland. Sequenced to an average depth of 20×, we recover 79% of the diploid genome, an amount close to the practical limit of current sequencing technologies. We identify 353,151 high-confidence single-nucleotide polymorphisms (SNPs), of which 6.8% have not been reported previously. We estimate raw read contamination to be no higher than 0.8%. We use functional SNP assessment to assign possible phenotypic characteristics of the individual that belonged to a culture whose location has yielded only trace human remains. We compare the high-confidence SNPs to those of contemporary populations to find the populations most closely related to the individual. This provides evidence for a migration from Siberia into the New World some 5,500 years ago, independent of that giving rise to the modern Native Americans and Inuit. For the first time, the sequence of a near-complete nuclear genome has been obtained from the tissue of an ancient human. It comes from permafrost-preserved hair, about 4,000 years old, of a male palaeo-Eskimo of the Saqqaq culture, the earliest known settlers in Greenland. Functional single-nucleotide polymorphism (SNP) assessment was used to assign possible phenotypic characteristics. The analysis provides evidence for a migration from Siberia into the New World some 5,500 years ago, independent of the migration that gave rise to the modern Native Americans and Inuit. Elsewhere in the issue we profile the paper's last author Eske Willerslev, who headed the project and found the lock of hair in a Copenhagen museum basement — after a fruitless search among the archaeological sites of Peary Land. The first genome sequence of an ancient human is reported. It comes from an approximately 4,000-year-old permafrost-preserved hair from a male from the first known culture to settle in Greenland. Functional single-nucleotide polymorphism (SNP) assessment is used to assign possible phenotypic characteristics and high-confidence SNPs are compared to those of contemporary populations to find those most closely related to the individual.

BACKGROUND: Statins increase the risk of new-onset type 2 diabetes mellitus. We aimed to assess whether this increase in risk is a consequence of inhibition of 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR), the intended drug target. METHODS: We used single nucleotide polymorphisms in the HMGCR gene, rs17238484 (for the main analysis) and rs12916 (for a subsidiary analysis) as proxies for HMGCR inhibition by statins. We examined associations of these variants with plasma lipid, glucose, and insulin concentrations; bodyweight; waist circumference; and prevalent and incident type 2 diabetes. Study-specific effect estimates per copy of each LDL-lowering allele were pooled by meta-analysis. These findings were compared with a meta-analysis of new-onset type 2 diabetes and bodyweight change data from randomised trials of statin drugs. The effects of statins in each randomised trial were assessed using meta-analysis. FINDINGS: Data were available for up to 223 463 individuals from 43 genetic studies. Each additional rs17238484-G allele was associated with a mean 0·06 mmol/L (95% CI 0·05-0·07) lower LDL cholesterol and higher body weight (0·30 kg, 0·18-0·43), waist circumference (0·32 cm, 0·16-0·47), plasma insulin concentration (1·62%, 0·53-2·72), and plasma glucose concentration (0·23%, 0·02-0·44). The rs12916 SNP had similar effects on LDL cholesterol, bodyweight, and waist circumference. The rs17238484-G allele seemed to be associated with higher risk of type 2 diabetes (odds ratio [OR] per allele 1·02, 95% CI 1·00-1·05); the rs12916-T allele association was consistent (1·06, 1·03-1·09). In 129 170 individuals in randomised trials, statins lowered LDL cholesterol by 0·92 mmol/L (95% CI 0·18-1·67) at 1-year of follow-up, increased bodyweight by 0·24 kg (95% CI 0·10-0·38 in all trials; 0·33 kg, 95% CI 0·24-0·42 in placebo or standard care controlled trials and -0·15 kg, 95% CI -0·39 to 0·08 in intensive-dose vs moderate-dose trials) at a mean of 4·2 years (range 1·9-6·7) of follow-up, and increased the odds of new-onset type 2 diabetes (OR 1·12, 95% CI 1·06-1·18 in all trials; 1·11, 95% CI 1·03-1·20 in placebo or standard care controlled trials and 1·12, 95% CI 1·04-1·22 in intensive-dose vs moderate dose trials). INTERPRETATION: The increased risk of type 2 diabetes noted with statins is at least partially explained by HMGCR inhibition. FUNDING: The funding sources are cited at the end of the paper.

Native Americans derive from a small number of Asian founders who likely arrived to the Americas via Beringia. However, additional details about the initial colonization of the Americas remain unclear. To investigate the pioneering phase in the Americas we analyzed a total of 623 complete mtDNAs from the Americas and Asia, including 20 new complete mtDNAs from the Americas and seven from Asia. This sequence data was used to direct high-resolution genotyping from 20 American and 26 Asian populations. Here we describe more genetic diversity within the founder population than was previously reported. The newly resolved phylogenetic structure suggests that ancestors of Native Americans paused when they reached Beringia, during which time New World founder lineages differentiated from their Asian sister-clades. This pause in movement was followed by a swift migration southward that distributed the founder types all the way to South America. The data also suggest more recent bi-directional gene flow between Siberia and the North American Arctic.

We review the evolution of domestic animals, emphasizing the effect of the earliest steps of domestication on its course. Using the first domesticated species, the dog (Canis familiaris), for illustration, we describe the evolutionary peculiarities during the historical domestication, such as the high level and wide range of diversity. We suggest that the process of earliest domestication via unconscious and later conscious selection of human-defined behavioral traits may accelerate phenotypic variations. The review is based on the results of a long-term experiment designed to reproduce early mammalian domestication in the silver fox (Vulpes vulpes) selected for tameability or amenability to domestication. We describe changes in behavior, morphology and physiology that appeared in the fox during its selection for tameability, which were similar to those observed in the domestic dog. Based on the data of the fox experiment and survey of relevant data, we discuss the developmental, genetic and possible molecular genetic mechanisms underlying these changes. We ascribe the causative role in evolutionary transformation of domestic animals to the selection for behavior and to the neurospecific regulatory genes it affects.

The sublethal effects of high predation risk on both prey behavior and physiology may have long-term consequences for prey population dynamics. We tested the hypothesis that snowshoe hares during the population decline are chronically stressed because of high predation risk whereas those during the population low are not, and that this has negative effects on both their physiology and demography. Snowshoe hares exhibit 10-yr population cycles; during declines, virtually every hare that dies is killed by a predator. We assessed the physiological responsiveness of the stress axis and of energy mobilization by subjecting hares during the population decline and low to a hormonal-challenge protocol. We monitored the population demography through live-trapping and assessed reproduction through a maternal-cage technique. During the 1990s' decline in the Yukon, Canada, hares were chronically stressed—as indicated by higher levels of free cortisol, reduced maximum corticosteroid-binding capacity, reduced testosterone response, reduced index of body condition, reduced leucocyte counts, increased overwinter body-mass loss, and increased glucose mobilization, relative to hares during the population low. This evidence is consistent with the explanation that predation risk, not high hare density or poor nutritional condition, accounted for the chronic stress and for the marked deterioration of reproduction during the decline. Reproduction and indices of stress physiology did not improve until predation risk declined. These findings may also account for the lag in recovery of hare reproduction after predator densities have declined and thus may implicate the long-term consequences of predation risk on prey populations beyond the immediate effects of predators on prey behavior and physiology.

How and when the Americas were populated remains contentious. Using ancient and modern genome-wide data, we found that the ancestors of all present-day Native Americans, including Athabascans and Amerindians, entered the Americas as a single migration wave from Siberia no earlier than 23 thousand years ago (ka) and after no more than an 8000-year isolation period in Beringia. After their arrival to the Americas, ancestral Native Americans diversified into two basal genetic branches around 13 ka, one that is now dispersed across North and South America and the other restricted to North America. Subsequent gene flow resulted in some Native Americans sharing ancestry with present-day East Asians (including Siberians) and, more distantly, Australo-Melanesians. Putative "Paleoamerican" relict populations, including the historical Mexican Pericúes and South American Fuego-Patagonians, are not directly related to modern Australo-Melanesians as suggested by the Paleoamerican Model.

Comparison of the genomes and proteomes of the two diptera Anopheles gambiae and Drosophila melanogaster, which diverged about 250 million years ago, reveals considerable similarities. However, numerous differences are also observed; some of these must reflect the selection and subsequent adaptation associated with different ecologies and life strategies. Almost half of the genes in both genomes are interpreted as orthologs and show an average sequence identity of about 56%, which is slightly lower than that observed between the orthologs of the pufferfish and human (diverged about 450 million years ago). This indicates that these two insects diverged considerably faster than vertebrates. Aligned sequences reveal that orthologous genes have retained only half of their intron/exon structure, indicating that intron gains or losses have occurred at a rate of about one per gene per 125 million years. Chromosomal arms exhibit significant remnants of homology between the two species, although only 34% of the genes colocalize in small "microsyntenic" clusters, and major interarm transfers as well as intra-arm shuffling of gene order are detected.

It is commonly thought that human genetic diversity in non-African populations was shaped primarily by an out-of-Africa dispersal 50-100 thousand yr ago (kya). Here, we present a study of 456 geographically diverse high-coverage Y chromosome sequences, including 299 newly reported samples. Applying ancient DNA calibration, we date the Y-chromosomal most recent common ancestor (MRCA) in Africa at 254 (95% CI 192-307) kya and detect a cluster of major non-African founder haplogroups in a narrow time interval at 47-52 kya, consistent with a rapid initial colonization model of Eurasia and Oceania after the out-of-Africa bottleneck. In contrast to demographic reconstructions based on mtDNA, we infer a second strong bottleneck in Y-chromosome lineages dating to the last 10 ky. We hypothesize that this bottleneck is caused by cultural changes affecting variance of reproductive success among males.

Importance: Blood pressure (BP) is a known risk factor for overall mortality and cardiovascular (CV)-specific fatal and nonfatal outcomes. It is uncertain which BP index is most strongly associated with these outcomes. Objective: To evaluate the association of BP indexes with death and a composite CV event. Design, Setting, and Participants: Longitudinal population-based cohort study of 11 135 adults from Europe, Asia, and South America with baseline observations collected from May 1988 to May 2010 (last follow-ups, August 2006-October 2016). Exposures: Blood pressure measured by an observer or an automated office machine; measured for 24 hours, during the day or the night; and the dipping ratio (nighttime divided by daytime readings). Main Outcomes and Measures: Multivariable-adjusted hazard ratios (HRs) expressed the risk of death or a CV event associated with BP increments of 20/10 mm Hg. Cardiovascular events included CV mortality combined with nonfatal coronary events, heart failure, and stroke. Improvement in model performance was assessed by the change in the area under the curve (AUC). Results: Among 11 135 participants (median age, 54.7 years, 49.3% women), 2836 participants died (18.5 per 1000 person-years) and 2049 (13.4 per 1000 person-years) experienced a CV event over a median of 13.8 years of follow-up. Both end points were significantly associated with all single systolic BP indexes (P < .001). For nighttime systolic BP level, the HR for total mortality was 1.23 (95% CI, 1.17-1.28) and for CV events, 1.36 (95% CI, 1.30-1.43). For the 24-hour systolic BP level, the HR for total mortality was 1.22 (95% CI, 1.16-1.28) and for CV events, 1.45 (95% CI, 1.37-1.54). With adjustment for any of the other systolic BP indexes, the associations of nighttime and 24-hour systolic BP with the primary outcomes remained statistically significant (HRs ranging from 1.17 [95% CI, 1.10-1.25] to 1.87 [95% CI, 1.62-2.16]). Base models that included single systolic BP indexes yielded an AUC of 0.83 for mortality and 0.84 for the CV outcomes. Adding 24-hour or nighttime systolic BP to base models that included other BP indexes resulted in incremental improvements in the AUC of 0.0013 to 0.0027 for mortality and 0.0031 to 0.0075 for the composite CV outcome. Adding any systolic BP index to models already including nighttime or 24-hour systolic BP did not significantly improve model performance. These findings were consistent for diastolic BP. Conclusions and Relevance: In this population-based cohort study, higher 24-hour and nighttime blood pressure measurements were significantly associated with greater risks of death and a composite CV outcome, even after adjusting for other office-based or ambulatory blood pressure measurements. Thus, 24-hour and nighttime blood pressure may be considered optimal measurements for estimating CV risk, although statistically, model improvement compared with other blood pressure indexes was small.

For pedigree-based quantitative trait loci (QTL) association analysis, a range of methods utilizing within-family variation such as transmission-disequilibrium test (TDT)-based methods have been developed. In scenarios where stratification is not a concern, methods exploiting between-family variation in addition to within-family variation, such as the measured genotype (MG) approach, have greater power. Application of MG methods can be computationally demanding (especially for large pedigrees), making genomewide scans practically infeasible. Here we suggest a novel approach for genomewide pedigree-based quantitative trait loci (QTL) association analysis: genomewide rapid association using mixed model and regression (GRAMMAR). The method first obtains residuals adjusted for family effects and subsequently analyzes the association between these residuals and genetic polymorphisms using rapid least-squares methods. At the final step, the selected polymorphisms may be followed up with the full measured genotype (MG) analysis. In a simulation study, we compared type 1 error, power, and operational characteristics of the proposed method with those of MG and TDT-based approaches. For moderately heritable (30%) traits in human pedigrees the power of the GRAMMAR and the MG approaches is similar and is much higher than that of TDT-based approaches. When using tabulated thresholds, the proposed method is less powerful than MG for very high heritabilities and pedigrees including large sibships like those observed in livestock pedigrees. However, there is little or no difference in empirical power of MG and the proposed method. In any scenario, GRAMMAR is much faster than MG and enables rapid analysis of hundreds of thousands of markers.

Fine structural details of glycans attached to the conserved N-glycosylation site significantly not only affect function of individual immunoglobulin G (IgG) molecules but also mediate inflammation at the systemic level. By analyzing IgG glycosylation in 5,117 individuals from four European populations, we have revealed very complex patterns of changes in IgG glycosylation with age. Several IgG glycans (including FA2B, FA2G2, and FA2BG2) changed considerably with age and the combination of these three glycans can explain up to 58% of variance in chronological age, significantly more than other markers of biological age like telomere lengths. The remaining variance in these glycans strongly correlated with physiological parameters associated with biological age. Thus, IgG glycosylation appears to be closely linked with both chronological and biological ages. Considering the important role of IgG glycans in inflammation, and because the observed changes with age promote inflammation, changes in IgG glycosylation also seem to represent a factor contributing to aging. SIGNIFICANCE STATEMENT: Glycosylation is the key posttranslational mechanism that regulates function of immunoglobulins, with multiple systemic repercussions to the immune system. Our study of IgG glycosylation in 5,117 individuals from four European populations has revealed very extensive and complex changes in IgG glycosylation with age. The combined index composed of only three glycans explained up to 58% of variance in age, considerably more than other biomarkers of age like telomere lengths. The remaining variance in these glycans strongly correlated with physiological parameters associated with biological age; thus, IgG glycosylation appears to be closely linked with both chronological and biological ages. The ability to measure human biological aging using molecular profiling has practical applications for diverse fields such as disease prevention and treatment, or forensics.

BACKGROUND: Over the last few years, genome-wide association (GWA) studies became a tool of choice for the identification of loci associated with complex traits. Currently, imputed single nucleotide polymorphisms (SNP) data are frequently used in GWA analyzes. Correct analysis of imputed data calls for the implementation of specific methods which take genotype imputation uncertainty into account. RESULTS: We developed the ProbABEL software package for the analysis of genome-wide imputed SNP data and quantitative, binary, and time-till-event outcomes under linear, logistic, and Cox proportional hazards models, respectively. For quantitative traits, the package also implements a fast two-step mixed model-based score test for association in samples with differential relationships, facilitating analysis in family-based studies, studies performed in human genetically isolated populations and outbred animal populations. CONCLUSIONS: ProbABEL package provides fast efficient way to analyze imputed data in genome-wide context and will facilitate future identification of complex trait loci.

In order to explore the diversity and selective signatures of duplication and deletion human copy-number variants (CNVs), we sequenced 236 individuals from 125 distinct human populations. We observed that duplications exhibit fundamentally different population genetic and selective signatures than deletions and are more likely to be stratified between human populations. Through reconstruction of the ancestral human genome, we identify megabases of DNA lost in different human lineages and pinpoint large duplications that introgressed from the extinct Denisova lineage now found at high frequency exclusively in Oceanic populations. We find that the proportion of CNV base pairs to single-nucleotide-variant base pairs is greater among non-Africans than it is among African populations, but we conclude that this difference is likely due to unique aspects of non-African population history as opposed to differences in CNV load.

Glycosylation of immunoglobulin G (IgG) influences IgG effector function by modulating binding to Fc receptors. To identify genetic loci associated with IgG glycosylation, we quantitated N-linked IgG glycans using two approaches. After isolating IgG from human plasma, we performed 77 quantitative measurements of N-glycosylation using ultra-performance liquid chromatography (UPLC) in 2,247 individuals from four European discovery populations. In parallel, we measured IgG N-glycans using MALDI-TOF mass spectrometry (MS) in a replication cohort of 1,848 Europeans. Meta-analysis of genome-wide association study (GWAS) results identified 9 genome-wide significant loci (P<2.27 × 10(-9)) in the discovery analysis and two of the same loci (B4GALT1 and MGAT3) in the replication cohort. Four loci contained genes encoding glycosyltransferases (ST6GAL1, B4GALT1, FUT8, and MGAT3), while the remaining 5 contained genes that have not been previously implicated in protein glycosylation (IKZF1, IL6ST-ANKRD55, ABCF2-SMARCD3, SUV420H1, and SMARCB1-DERL3). However, most of them have been strongly associated with autoimmune and inflammatory conditions (e.g., systemic lupus erythematosus, rheumatoid arthritis, ulcerative colitis, Crohn's disease, diabetes type 1, multiple sclerosis, Graves' disease, celiac disease, nodular sclerosis) and/or haematological cancers (acute lymphoblastic leukaemia, Hodgkin lymphoma, and multiple myeloma). Follow-up functional experiments in haplodeficient Ikzf1 knock-out mice showed the same general pattern of changes in IgG glycosylation as identified in the meta-analysis. As IKZF1 was associated with multiple IgG N-glycan traits, we explored biomarker potential of affected N-glycans in 101 cases with SLE and 183 matched controls and demonstrated substantial discriminative power in a ROC-curve analysis (area under the curve = 0.842). Our study shows that it is possible to identify new loci that control glycosylation of a single plasma protein using GWAS. The results may also provide an explanation for the reported pleiotropy and antagonistic effects of loci involved in autoimmune diseases and haematological cancer.

The three Drosophila EcR isoforms differ only at their N termini; thus, they share the conserved ligand-binding domain transcriptional activation function (AF2) and only differ in the unconserved A/B region, which contains a second, isoform-specific, activation function (AF1). We have developed a dominant-negative mutant EcR (EcR-DN), expressed it in flies with the GAL4/UAS system, and used it to block ecdysone signaling in eight tissues or groups of tissues. Localized EcR-DN arrests ecdysone-dependent development in the target cells and often--because of a molting checkpoint--arrests development globally. Simultaneously expressing individual wild-type EcR isoforms in the same target tissues suppresses the EcR-DN phenotype and identifies the rescuing isoform as sufficient to support the development of the target. Every isoform, and even an N-terminal truncated EcR that lacks any AF1, supports development in the fat body, eye discs, salivary glands, EH-secreting neurosecretory cells and in the dpp expression domain, implying that AF1 is dispensable in these tissues. By contrast, only EcR-A is able to support development in the margins of the wing discs, and only EcR-B2 can do so in the larval epidermis and the border cells of the developing egg chamber. In light of our results, the simplest explanations for the widespread spatial and temporal variations in EcR isoform titers appear untenable.

We studied the interactions of phosphorothioate oligodeoxynucleotides and heparin-binding growth factors. By means of a gel mobility shift assay, we demonstrated that phosphodiester and phosphorothioate homopolymers bound to basic fibroblast growth factor (bFGF). Binding of a probe phosphodiester oligodeoxynucleotide could also be shown for other proteins of the FGF family, including acidic fibroblast growth factor (aFGF), Kaposi's growth factor (FGF-4) as well as for the bFGF-related vascular endothelial growth factor, VEGF. No binding to epidermal growth factor (EGF) was observed. In addition, using a radioreceptor assay, we have shown that phosphorothioate homopolymers of cytidine and thymidine blocked binding of not only 125I-bFGF, but also of 125I-PDGF to NIH 3T3 cells, whereas phosphodiester oligodeoxynucleotides were ineffective. The extent of blockade of binding was dependent on the chain length of the phosphorothioate oligodeoxynucleotide. Furthermore, we have examined the effects of 18-mer phosphorothioate oligodeoxynucleotides of different sequences on 125I-bFGF binding to low and high affinity sites on both NIH 3T3 fibroblasts and DU-145 prostate cancer cells. Despite the fact that we have observed inhibition of bFGF binding by the 18-mer phosphorothioate oligodeoxynucleotides for both the high and low affinity classes of bFGF receptor, the inhibition was sequence-selective only for the high affinity receptors. We have also demonstrated that phosphorothioate homopolymers of cytidine and thymidine release bFGF bound to low affinity receptors in extracellular matrix (ECM). Finally, the most potent phosphorothioate oligodeoxynucleotides used in these experiments (e.g. SdC28) were inhibitors of bFGF-induced DNA synthesis in NIH 3T3 cells. We studied the interactions of phosphorothioate oligodeoxynucleotides and heparin-binding growth factors. By means of a gel mobility shift assay, we demonstrated that phosphodiester and phosphorothioate homopolymers bound to basic fibroblast growth factor (bFGF). Binding of a probe phosphodiester oligodeoxynucleotide could also be shown for other proteins of the FGF family, including acidic fibroblast growth factor (aFGF), Kaposi's growth factor (FGF-4) as well as for the bFGF-related vascular endothelial growth factor, VEGF. No binding to epidermal growth factor (EGF) was observed. In addition, using a radioreceptor assay, we have shown that phosphorothioate homopolymers of cytidine and thymidine blocked binding of not only 125I-bFGF, but also of 125I-PDGF to NIH 3T3 cells, whereas phosphodiester oligodeoxynucleotides were ineffective. The extent of blockade of binding was dependent on the chain length of the phosphorothioate oligodeoxynucleotide. Furthermore, we have examined the effects of 18-mer phosphorothioate oligodeoxynucleotides of different sequences on 125I-bFGF binding to low and high affinity sites on both NIH 3T3 fibroblasts and DU-145 prostate cancer cells. Despite the fact that we have observed inhibition of bFGF binding by the 18-mer phosphorothioate oligodeoxynucleotides for both the high and low affinity classes of bFGF receptor, the inhibition was sequence-selective only for the high affinity receptors. We have also demonstrated that phosphorothioate homopolymers of cytidine and thymidine release bFGF bound to low affinity receptors in extracellular matrix (ECM). Finally, the most potent phosphorothioate oligodeoxynucleotides used in these experiments (e.g. SdC28) were inhibitors of bFGF-induced DNA synthesis in NIH 3T3 cells. INTRODUCTIONPhosphorothioate oligodeoxynucleotides are isoelectronic congeners of phosphodiester oligodeoxynucleotides that retain the property of aqueous solubility and Watson-Crick base pair hybridization, but which are also nuclease-resistant (Stein et al., 1988). These materials have found wide application as both in vitro and in vivo sequence-specific, or antisense, inhibitors of gene expression (for a review, see Stein and Cheng(1993)). However, it has been recognized for some years that these compounds may have non-sequence-specific effects on cellular function. These may result, at least in part, from their ability to bind to cellular proteins. For example, phosphorothioate oligodeoxynucleotides appear to bind non-sequence specifically to rsCD4 (Yakubov et al., 1993), gp120 (Stein et al., 1993), and to protein kinase C β1, α, δ, and ε isoforms. Other polyanions, including pentosan polysulfate (Wellstein et al., 1991) and suramin (Stein, 1993), can also bind to these proteins. Furthermore, these latter polyanions also bind to heparin-binding growth factors, including bFGF ( 1The abbreviations used are: bFGFbasic fibroblast growth factoraFGFacidic fibroblast growth factorFGF-4fibroblast growth factor-4PDGFplatelet-derived growth factorVEGFvascular endothelial growth factorEGFepidermal growth factorSdC2828-mer phosphorothioate homopolymer of cytidineECMextracellular matrixBSAbovine serum albuminDMEMDulbecco's modified Eagle's mediumDPBSDulbecco's phosphate-buffered salinePAGEpolyacrylamide gel electrophoresis. )(Moscatelli and Quarto, 1989) and other growth factors as well (Coffey et al., 1987). We hypothesized that oligodeoxynucleotides, which are also polyanions, might, similar to suramin and pentosan polysulfate, interact with bFGF and other heparin-binding proteins. In this report, we demonstrate direct binding of a phosphodiester probe oligodeoxynucleotide to bFGF by means of a mobility shift assay in denaturing polyacrylamide gels. We also demonstrate that phosphorothioate oligodeoxynucleotides can block the binding of human 125I-labeled bFGF to both low and high affinity receptors on the surface of NIH 3T3 and DU-145 cells. We further show that phosphorothioate oligodeoxynucleotides, similar to heparin, can remove 125I-labeled bFGF from its low affinity binding sites on subendothelial extracellular matrix. In addition, on the basis of our data, we suggest that the ability of phosphorothioate oligodeoxynucleotides to block the binding of 125I-labeled bFGF to its cell surface receptors may be sequence-selective.MATERIALS AND METHODSReagents125I-Bolton-Hunter-labeled bFGF (human, recombinant), 125I-Bolton-Hunter-labeled PDGF, and 125I-EGF (murine) were obtained from NEN Research Products. Unlabeled EGF, VEGF, aFGF, bFGF, and FGF-4 were obtained from R and D Systems (Minneapolis, MN). Basic fibroblast growth factor (human, recombinant) was also obtained from Promega or as a generous gift from Takeda Chemical Industries (Osaka, Japan). Bovine serum albumin (BSA) was purchased from Sigma. Heparin sodium, Mr = 14,000, was obtained from Hepar Industries (Franklin, OH).Synthesis of OligodeoxynucleotidesPhosphodiester oligodeoxynucleotides were synthesized by standard phosphoramidite chemistry on an Applied Biosystems 380B synthesizer. Phosphorothioate oligodeoxynucleotides were also synthesized by standard methods (Stein et al., 1988), and sulfurization was performed using the tetraethylthiuram disulfide/acetonitrile reagent (TETD; Applied Biosystems). Following cleavage from the controlled glass support, oligodeoxynucleotides were base-deblocked in ammonium hydroxide at 60°C for 8 h and purified by reversed phase high performance liquid chromatography (0.1 M triethylammonium bicarbonate/acetonitrile, PRP-1 support). Oligomers were detritylated in 3% acetic acid and precipitated with 2% lithium perchlorate/acetone dissolved in sterile water and reprecipitated as the sodium salt from 1 M NaCl/ethanol. Oligodeoxynucleotide concentrations were determined by UV spectroscopy.In addition to phosphorothioate homopolymers of cytidine and thymidine, we also used five 18-heteromer oligodeoxynucleotides of different sequences to block binding of bFGF to its cell surface receptors. Three oligodeoxynucleotides (1, 2, and 3) were complementary to codons 2-7 of either the rat or mouse c-myb mRNA. In addition, one of the oligodeoxynucleotides (No. 3; antisense rat c-myb) was a chimeric phosphorothioate/diester (two phosphorothioate linkages at the 5′ and five phosphorothioate linkages at the 3′ terminus). One oligodeoxynucleotide (No. 4) was sense rat c-myb, and the other (No. 5) was a scrambled version of the rat antisense c-myb oligodeoxynucleotide. The sequences are: 1, 5′-GTGCCGGGGTCTCCGGGC-3′ (antisense rat c-myb, all phosphorothioate); 2, 5′-GTGTCGGGGTCTCCGGGC-3′ (antisense mouse c-myb, all phosphorothioate); 3, 5′-GSTSGCCGGGGTCTCSCSGSGSGSC-3′ (chimeric phosphorothioate/diester); 4, 5′-GCCCGGAGACCCCGGCAC-3′ (sense rat c-myb, all phosphorothioate); 5, 5′-CGCCGTCGCGGCGGTTGG-3′ (scrambled rat c-myb, all phosphorothioate).Synthesis of Alkylating, Radioactive Phosphodiester Oligodeoxynucleotide 5′-N-Methyl-N-(2-chloroethyl)aminobenzylamine-32P-OdT18 or -32P-OdT12 (RClNH32P-OdT18 or RClNH32P-OdT12)This compound was synthesized by a modification of the method of Knorre et al.(1985). Briefly, after 5′-phosphorylation of OdT18 or OdT12 by Chemical Phosphorylation Reagent (Glen Research, Herndon, VA), a reaction exchanging the 32P of [γ-32P]ATP was carried out using T4 polynucleotide kinase with ADP as the phosphate acceptor (Sambrook et al., 1989). Then, N-methyl-N-(2-chloroethyl)aminobenzylamine was coupled to the 5′-terminal 32P by reaction with triphenylphosphine/dipyridyl disulfide. The final product was stored at −70°C.Modification of bFGF and Other Heparin-binding Growth Factors by ClRNH32P-OdT18 or ClRNH32P-OdT12This was accomplished by the method of Yakubov et al.(1993). bFGF (10 μg/ml), EGF (3 μg/ml), aFGF (10 μg/ml), FGF4 (10 μg/ml), or VEGF (25 μg/ml) was incubated in 0.1 M Tris-HCl, pH 7.6, containing the appropriate concentration of ClRNH32P-OdT18 or ClRNH32P-OdT12. ( 2ClRNH32P-OdT12 is dodecathymidylate phosphodiester oligodeoxynucleotide derivative with an alkylator moiety (Fig. ZI) coupled to the 5′ radioactive phosphate through a phosphoroamide some a (e.g. phosphorothioate of the binding of the to bFGF was also as in the 1 of a containing 0.1 M 2% and was and was The were and to were The was and were by NIH 3T3 and the human prostate were obtained from NIH 3T3 were and in modified Eagle's containing serum and DU-145 were and in containing serum sodium, and For binding were at 1 well in or 1 well in and were used for experiments 1 to of endothelial were from as et al., were in of with and at in purified bFGF was other the phase of cell growth et al., et al., of 125I-bFGF to of were used for all were with phosphate-buffered and incubated in with at for 1 h the of binding For the binding were incubated at in containing 125I-bFGF, and the concentrations of the of the the was and 125I-bFGF was by one of bound bFGF, the were with and were with 1 M in other were with and with M in pH to remove 125I-bFGF bound to the low affinity binding the were with in 0.1 M sodium pH The binding to high affinity sites 1987). binding was in the of of The of in the and was determined in a 1, of with endothelial were from to with M sodium phosphate and and at an of were as that was in the growth and the were for addition of bFGF et al., et al., 1987). The subendothelial was by the cell with containing and by in The of cellular and to the of the et al., et al., of was incubated (3 with 125I-bFGF in containing bFGF was and the was incubated with concentrations of either or the oligodeoxynucleotide at for The were and in a to the of The was incubated (3 with 1 and the was in a The of 125I-bFGF was from the et al., on the of of the bound 125I-bFGF was with release could be by the to its release was from the to the specifically was performed to five similar 3T3 were at a concentration of well in were at the was and were with containing The was with with bFGF, and the appropriate concentration of phosphorothioate oligodeoxynucleotides and were for h with The of was on glass using a liquid of bFGF by the bFGF and the were incubated in 0.1 M for 1 and the reaction were to in a a was observed on the to the product of modification of the protein by the of the as oligodeoxynucleotide of not containing an moiety not with bFGF, in this gel were on the No in the modification of bFGF by could be in different including 0.1 M in M 0.1 M or in examined the concentration of the modification of bFGF by ClRNH32P-OdT18 (Fig. These are also in the concentration of oligodeoxynucleotide is as a of gel as determined by The of bFGF with the probe oligodeoxynucleotide the of the in is not and the fact that it is that the concentration of bFGF modification be by the However, at low concentrations of the concentration of the modification is and the to an of concentrations the of modification is also and the at = These that are at least binding with different for on the surface of the bFGF of bFGF by the oligodeoxynucleotide bFGF (3 μg/ml) was incubated in 0.1 M with ClRNH32P-OdT18 at the concentration for at The containing the was to polyacrylamide gel electrophoresis. The concentration of ClRNH32P-OdT18 was as 2, and ClRNH32P-OdT18 the of the gel and is not concentration of modification of bFGF by concentrations of The gel in 1 were by is a of oligodeoxynucleotide concentration The was of the in The of the in The of the with the were to this binding we used a gel to the binding of ClRNH32P-OdT18 to gel is shown in 3, and are the other bFGF a at the appropriate after The the of the gel at the for a bFGF of bFGF by the oligodeoxynucleotide ClRNH32P-OdT18 with gel performed by bFGF (3 μg/ml) was incubated in 0.1 M with ClRNH32P-OdT18 as in the The concentration of ClRNH32P-OdT18 was 1, and probe is at the of the are which at the of bFGF = The the of the gel at the for a bFGF of the for a of Oligodeoxynucleotide Binding to have examined the ability of other polyanions, including the phosphodiester oligodeoxynucleotides and to with the probe oligodeoxynucleotide for binding to of these polyanions are of binding to bFGF of the probe oligodeoxynucleotide of probe = Furthermore, a (Stein, to be to bind to heparin-binding growth factors and to block their binding to cell surface also is a of the binding of to ( and in examined the ability of a phosphorothioate homopolymer of thymidine, to binding of the oligodeoxynucleotide to We have used this method to the of for of oligodeoxynucleotide binding to The of may be from 1 from and = the we found (Fig. 4, and that the of = However, the of is by the in a of is the of the low affinity and the high affinity of the oligodeoxynucleotide binding to bFGF are and are in the of we have used an for the of The of for as determined by the is by for binding of to was used as a of binding to bFGF as in the bFGF modified by the oligodeoxynucleotide. The concentration of was and of the of by the The from the was by is a = of the of the concentration The was of to Other Heparin-binding Growth was performed in a gel and is shown in can be not only bFGF, but acidic FGF4 and VEGF can bind to either of the probe oligodeoxynucleotides or concentration of = In to the with these growth factors, is at the of of EGF, which to only be that the of aFGF, and VEGF all has been demonstrated et al., that albumin can bind phosphodiester oligodeoxynucleotides with low The modification of the albumin in these by can be at the in of growth factors by ClRNH32P-OdT18 or The growth factor was incubated with either or 1 and EGF (3 and bFGF (10 and aFGF (10 and FGF4 (10 and VEGF (25 of the FGF at the 1 on the VEGF at the No binding of either probe oligodeoxynucleotide to EGF was observed. In bFGF, aFGF, and VEGF all Other in and protein The by the modification of which is to the reagent by the of Growth Binding to by to oligodeoxynucleotides can interact with growth factors and can their binding to cell surface we examined the binding of basic FGF to 3T3 fibroblasts in the of different phosphodiester and phosphorothioate homopolymers of cytidine and thymidine and were for their ability to 125I-labeled bFGF binding (Fig. a of oligodeoxynucleotide concentrations and we observed inhibition of bFGF binding by phosphodiester However, phosphorothioate homopolymers of thymidine as well as cytidine of of phosphorothioate and phosphodiester homopolymers of cytidine and thymidine on binding of 125I-bFGF to NIH 3T3 cells. were in at For were incubated at in containing 125I-bFGF, and the concentrations of was and were with phosphate-buffered and with 1 M was determined by are the of by of the binding assay, we demonstrated that the inhibition of bFGF binding in with the oligodeoxynucleotide chain length for homopolymers of cytidine and the the oligodeoxynucleotide or the its at concentration (Fig. We also examined the ability of phosphorothioate oligodeoxynucleotides to the binding of heparin-binding growth factor to cells. In a binding assay using 125I-labeled PDGF, we demonstrated that phosphorothioate homopolymers of cytidine of different chain length binding to the inhibition is chain with and least and most (Fig. the other these phosphorothioate oligodeoxynucleotides not binding of EGF to cell surface receptors of 3T3 (Fig. length dependent effects of phosphorothioate homopolymers of cytidine and thymidine on binding of 125I-bFGF to NIH 3T3 cells. were in at For were incubated at in containing 125I-bFGF, and the concentrations of was and were with phosphate-buffered and with 1 M was determined by are the of of phosphorothioate homopolymers of cytidine on binding of 125I-PDGF and 125I-EGF to NIH 3T3 cells. were in at For were incubated at in containing or 0.1 125I-EGF and the concentrations of the was and were with phosphate-buffered and with 1 M was determined by is the of which by of bFGF Binding to and by Oligodeoxynucleotide bFGF binding to low and high affinity we used method of 18-mer phosphorothioate oligodeoxynucleotides the binding of bFGF to the low affinity receptors of fibroblasts with similar (Fig. However, in oligodeoxynucleotide were found the ability to high affinity binding of bFGF was determined (Fig. In these the concentrations of oligodeoxynucleotides were (Fig. the oligodeoxynucleotide concentration the high affinity binding of bFGF was by in the of both the antisense c-myb rat and mouse 1 and In binding was only by by the scrambled phosphorothioate oligodeoxynucleotide (No. and by only in the of chimeric phosphorothioate/diester oligodeoxynucleotide (No. Furthermore, we to of the sense oligodeoxynucleotide (No. 4) on high affinity binding at the concentrations were high and low affinity binding of bFGF to the prostate cancer cell DU-145 were (Fig. and of 18-mer phosphorothioate oligodeoxynucleotides on binding of to low and high affinity receptors on cells. were in at For binding were incubated at in containing 125I-bFGF, and the concentrations of was and bFGF bound to low and high affinity receptors was determined as and are the of Oligodeoxynucleotide are and oligodeoxynucleotide oligodeoxynucleotide oligodeoxynucleotide 3; oligodeoxynucleotide oligodeoxynucleotide of 18-mer oligodeoxynucleotides on binding of 125I-bFGF to low and high affinity receptors on prostate cancer cells. were in at For were incubated at in containing 125I-bFGF, and the concentrations of was and bFGF bound to low and high affinity receptors was determined as and are the of Oligodeoxynucleotide are and oligodeoxynucleotide oligodeoxynucleotide oligodeoxynucleotide 3; oligodeoxynucleotide oligodeoxynucleotide are of of bFGF by Phosphorothioate on the of bFGF with the subendothelial have shown that bFGF to in the extracellular matrix and can be by and et al., In the was release of bFGF by a phosphodiester homopolymer containing to a concentration of In of extracellular matrix to phosphorothioate homopolymers of thymidine and cytidine in release of release was dependent on oligodeoxynucleotide chain for example, bound bFGF, whereas of the bound bFGF (Fig. was as as release at but with release at Phosphorothioate of cytidine were the thymidine congeners of the length release at for release of of the extracellular bFGF was obtained in the of a concentration of both and of bFGF by homopolymer phosphorothioate oligodeoxynucleotides of different chain of were incubated (3 with 125I-bFGF The was and incubated (3 with concentrations of and is as the of 125I-bFGF of of 125I-bFGF in the of oligodeoxynucleotides not of the is the of and the standard not other we examined the ability of 18-heteromer oligodeoxynucleotides 1, 2, and 4) to release extracellular bFGF was by of these a concentration of 1 the of bFGF was for the antisense c-myb mouse compound and for the antisense and sense rat c-myb a similar the release for and was and of bFGF-induced by Phosphorothioate used bFGF as a of fibroblast We the effects of phosphorothioate oligodeoxynucleotides on the ability of the to These oligodeoxynucleotides phosphorothioate homopolymers of cytidine and and oligodeoxynucleotides 1, 2, and as The addition of compounds 1 and 2, as well as in a The most effects were observed a of concentrations In the sense was to the extent (Fig. In with as cell and were of phosphorothioate oligodeoxynucleotides on by NIH 3T3 cells. were for h in containing bFGF only or phosphorothioate oligodeoxynucleotides, as was a are of fibroblast growth factors of a of at least effects are through a of high affinity receptors = but also interact with affinity receptors = et al., Despite the of a both aFGF and bFGF have also been in the extracellular matrix by and 1989) and endothelial et al., 1987). the of bFGF in of the rat et al., et al., 1988), and human et al., that may as a for bFGF et al., that bFGF specifically to and in the and cell as by its by heparin, or but not by or et al., et al., 1989). These may bFGF from can also be by the = et al., release from the the growth may at is to suggest that may to the et al., By a gel mobility shift assay, et have demonstrated that aFGF can with phosphodiester In this we have demonstrated that both phosphorothioate and phosphodiester oligodeoxynucleotides are of binding to In the of phosphorothioate oligodeoxynucleotides this binding may release of bFGF from its low affinity binding sites on extracellular matrix. Furthermore, phosphorothioate oligodeoxynucleotides can block the binding of bFGF to both low and high affinity cell surface receptors and can the bFGF-induced in in 3T3 ability of oligodeoxynucleotides to bind to growth factors is not to These compounds can also interact with aFGF, VEGF, and not with the basis of these data, it is not to that oligodeoxynucleotides may be to interact with heparin-binding growth factors. of oligodeoxynucleotides is of that of other polyanions, including suramin (Stein, et al., et al., et al., 1991) and pentosan polysulfate (Wellstein et al., et al., is a and a to bFGF and VEGF, PDGF, and to other proteins as only some of which are binding (Stein et al., least in to these suramin has and effects and is in cancer (Stein et al., et al., The in the of phosphorothioate oligodeoxynucleotides to that of suramin that the may also have sequence-selective and is in the addition to growth factors, are other proteins to which phosphorothioate oligodeoxynucleotides, suramin and pentosan polysulfate, can the studied of these is rsCD4 (Yakubov et al., In this binding sites have been to both the and of the In the bFGF are basic that are on its surface et al., the basis of a et al., et al., it has been that the of a bound be by the of in addition to the chain of binding also to and the chain of as well as the of and may be this was in demonstrated that for the binding of to bFGF et al., it is that phosphorothioate oligodeoxynucleotides also bind to bFGF at or the heparin-binding However, the and heparin-binding sites of bFGF appear to be et al., as shown by the fact that that the binding of bFGF to its not block it has been that as suramin phosphorothioate oligodeoxynucleotides as to binding to bFGF, either to the binding similar to and other et al., a in the bFGF et al., to our the for the binding of some of the to bFGF may be as low as Furthermore, some in a can block the binding of bFGF to high affinity receptors with an = 1 et al., However, of this by with phosphorothioate oligodeoxynucleotides of their to is that the ability of phosphorothioate oligodeoxynucleotides to block the binding of 125I-bFGF to cell surface receptors is at least has to experiments in which phosphorothioate oligodeoxynucleotides are to complementary on mRNA. In some experiments et al., it is that the antisense and the may ability to block the binding of heparin-binding growth factors to their cell surface receptors. determined to be a direct of Watson-Crick base pair may be to the non-sequence-specific effects of phosphorothioate oligodeoxynucleotides may at similar concentrations as the the may not be suggest that the the and the its the it is that observed effects have a non-sequence-specific is by our on the effects of phosphorothioate oligodeoxynucleotides on in 3T3 it be to that antisense oligodeoxynucleotide not interact the concentrations with the protein product of the (Stein and the we also that phosphorothioate homopolymers of thymidine and cytidine are of has been demonstrated that containing as as of bFGF of release of bFGF was by a et al., was or release of bFGF by and of the other was by of heparin, but was not by containing et al., In the for the phosphorothioate release of bFGF and for inhibition of were ( and suggest that the oligodeoxynucleotides their effects by of their the effects of are to a of the bFGF and In this phosphorothioate oligodeoxynucleotides to the of and may release bFGF from its in phosphorothioate oligodeoxynucleotides may bFGF that is from and cell and its in and These are also INTRODUCTIONPhosphorothioate oligodeoxynucleotides are isoelectronic congeners of phosphodiester oligodeoxynucleotides that retain the property of aqueous solubility and Watson-Crick base pair hybridization, but which are also nuclease-resistant (Stein et al., 1988). These materials have found wide application as both in vitro and in vivo sequence-specific, or antisense, inhibitors of gene expression (for a review, see Stein and Cheng(1993)). However, it has been recognized for some years that these compounds may have non-sequence-specific effects on cellular function. These may result, at least in part, from their ability to bind to cellular proteins. For example, phosphorothioate oligodeoxynucleotides appear to bind non-sequence specifically to rsCD4 (Yakubov et al., 1993), gp120 (Stein et al., 1993), and to protein kinase C β1, α, δ, and ε isoforms. Other polyanions, including pentosan polysulfate (Wellstein et al., 1991) and suramin (Stein, 1993), can also bind to these proteins. Furthermore, these latter polyanions also bind to heparin-binding growth factors, including bFGF ( 1The abbreviations used are: bFGFbasic fibroblast growth factoraFGFacidic fibroblast growth factorFGF-4fibroblast growth factor-4PDGFplatelet-derived growth factorVEGFvascular endothelial growth factorEGFepidermal growth factorSdC2828-mer phosphorothioate homopolymer of cytidineECMextracellular matrixBSAbovine serum albuminDMEMDulbecco's modified Eagle's mediumDPBSDulbecco's phosphate-buffered salinePAGEpolyacrylamide gel electrophoresis. )(Moscatelli and Quarto, 1989) and other growth factors as well (Coffey et al., 1987). We hypothesized that oligodeoxynucleotides, which are also polyanions, might, similar to suramin and pentosan polysulfate, interact with bFGF and other heparin-binding proteins. In this report, we demonstrate direct binding of a phosphodiester probe oligodeoxynucleotide to bFGF by means of a mobility shift assay in denaturing polyacrylamide gels. We also demonstrate that phosphorothioate oligodeoxynucleotides can block the binding of human 125I-labeled bFGF to both low and high affinity receptors on the surface of NIH 3T3 and DU-145 cells. We further show that phosphorothioate oligodeoxynucleotides, similar to heparin, can remove 125I-labeled bFGF from its low affinity binding sites on subendothelial extracellular matrix. In addition, on the basis of our data, we suggest that the ability of phosphorothioate oligodeoxynucleotides to block the binding of 125I-labeled bFGF to its cell surface receptors may be