Comparative mapping of human alphoid sequences in great apes using fluorescence in situ hybridization

CENP-C binds the alpha-satellite DNA in vivo at specific centromere domains

CENP-C is a fundamental component of the centromere, highly conserved among species and necessary for the proper assembly of the kinetochore structure and for the metaphase-anaphase transition. Although CENP-C can bind DNA in vitro,the identification of the DNA sequences associated with it in vivo and the significance of such an interaction have been, until now, elusive. To address this problem we took advantage of a chromatin-immunoprecipitation procedure and applied this technique to human HeLa cells. Through this approach we could establish that: (1) CENP-C binds the alpha-satellite DNA selectively; (2) the CENP-C region between amino acids 410 and 537, previously supposed to contain a DNA-binding domain, is indeed required to perform such a function in vivo;and (3) the profile of the alpha-satellite DNA associated with CENP-C is essentially identical to that recognized by CENP-B. However, further biochemical and ultrastructural characterization of CENP-B/DNA and CENP-C/DNA complexes, relative to their DNA components and specific spatial distribution in interphase nuclei, surprisingly reveals that CENP-C and CENP-B associate with the same types of alpha-satellite arrays but in distinct non-overlapping centromere domains. Our results, besides extending previous observations on the role of CENP-C in the formation of active centromeres, show, for the first time, that CENP-C can associate with the centromeric DNA sequences in vivo and, together with CENP-B, defines a highly structured organization of the alpha-satellite DNA within the human centromere.

Human chromosomes 9, 12, and 15 contain the nucleation sites of stress-induced nuclear bodies

We previously reported the identification of a novel nuclear compartment detectable in heat-shocked HeLa cells that we termed stress-induced Src-activated during mitosis nuclear body (SNB). This structure is the recruitment center for heat shock factor 1 and for a number of RNA processing factors, among a subset of Serine-Arginine splicing factors. In this article, we show that stress-induced SNBs are detectable in human but not in hamster cells. By means of hamster>human cell hybrids, we have identified three human chromosomes (9, 12, and 15) that are individually able to direct the formation of stress bodies in hamster cells. Similarly to stress-induced SNB, these bodies are sites of accumulation of hnRNP A1-interacting protein and heat shock factor 1, are usually associated to nucleoli, and consist of clusters of perichromatin granules. We show that the p13-q13 region of human chromosome 9 is sufficient to direct the formation of stress bodies in hamster>human cell hybrids. Fluorescence in situ hybridization experiments demonstrate that the pericentromeric heterochromatic q12 band of chromosome 9 and the centromeric regions of chromosomes 12 and 15 colocalize with stress-induced SNBs in human cells. Our data indicate that human chromosomes 9, 12, and 15 contain the nucleation sites of stress bodies in heat-shocked HeLa cells.

In vivo binding of active heat shock transcription factor 1 to human chromosome 9 heterochromatin during stress

Activation of the mammalian heat shock transcription factor (HSF)1 by stress is a multistep process resulting in the transcription of heat shock genes. Coincident with these events is the rapid and reversible redistribution of HSF1 to discrete nuclear structures termed HSF1 granules, whose function is still unknown. Key features are that the number of granules correlates with cell ploidy, suggesting the existence of a chromosomal target. Here we show that in humans, HSF1 granules localize to the 9q11-q12 heterochromatic region. Within this locus, HSF1 binds through direct DNA-protein interaction with a nucleosome-containing subclass of satellite III repeats. HSF1 granule formation only requires the DNA binding competence and the trimerization of the factor. This is the first example of a transcriptional activator that accumulates transiently and reversibly on a chromosome-specific heterochromatic locus.

UBF binding in vivo is not restricted to regulatory sequences within the vertebrate ribosomal DNA repeat

The HMG box containing protein UBF binds to the promoter of vertebrate ribosomal repeats and is required for their transcription by RNA polymerase I in vitro. UBF can also bind in vitro to a variety of sequences found across the intergenic spacer in Xenopus and mammalian ribosomal DNA (rDNA) repeats. The high abundance of UBF, its colocalization with rDNA in vivo, and its DNA binding characteristics, suggest that it plays a more generalized structural role over the rDNA repeat. Until now this view has not been supported by any in vivo data. Here, we utilize chromatin immunoprecipitation from a highly enriched nucleolar chromatin fraction to show for the first time that UBF binding in vivo is not restricted to known regulatory sequences but extends across the entire intergenic spacer and transcribed region of Xenopus, human, and mouse rDNA repeats. These results are consistent with a structural role for UBF at active nucleolar organizer regions in addition to its recognized role in stable transcription complex formation at the promoter.

Reconstruction of the female Gorilla gorilla karyotype using 25-color FISH and multicolor banding (MCB)

The origin of the human and great ape chromosomes has been studied by comparative chromosome banding analysis and, more recently, by fluorescence in situ hybridization (FISH), using human whole-chromosome painting probes. It is not always possible, however, to determine the exact breakpoints and distribution or orientation of specific DNA regions using these techniques. To overcome this problem, the recently developed multicolor banding (MCB) probe set for all human chromosomes was applied in the present study to reanalyze the chromosomes of Gorilla gorilla (GGO). While the results agree with those of most previous banding and FISH studies, the breakpoints for the pericentric inversion on GGO 3 were defined more precisely. Moreover, no paracentric inversion was found on GGO 14, and no pericentric inversions could be demonstrated on GGO 16 or 17.

Additional dark G-band in the p-arm of chromosome 19 due to a paracentric inversion with a breakpoint in the pericentromeric heterochromatin

Paracentric inversions in chromosome 19 have rarely been described. Here we present an inv(19)(p11p13.1) with a breakpoint in the pericentromeric heterochromatin which leads to an additional dark G-band in the p-arm of chromosome 19. The rearranged chromosome segregated in two generations of a family without any phenotypic effects. A detailed characterization of the inv(19) by molecular cytogenetic techniques is presented.

A novel translocation t(2;9)(q14;p12) in AML-M2 with an uncommon phenotype: myeloperoxidase-positive and myeloid antigen-negative

We report a case of acute myeloid leukemia (AML-M2) expressing myeloperoxidase (MPO) but no myeloid antigens. A few cases with this discordant phenotype have been reported and an association has been suggested between the lack of CD13 and CD33 in MPO positive AML and the presence of t(8;21). Cytogenetic and molecular analyses performed in our case showed 48,XY,+Y,+8,t(2;9)(q14;p12). We believe that combined approaches can contribute to detect particular AL cases like the present one, that confirms the heterogeneity of AML. However, further studies are needed to clarify the relationship between phenotypic aberrations and cytogenetic abnormalities.

The species and chromosomal distribution of the centromeric alpha-satellite I sequence from sheep in the tribe Caprini and other Bovidae

The evolution of chromosomes in species in the family Bovidae includes fusion and fission of chromosome arms (giving different numbers of acrocentric and metacentric chromosomes with a relatively conserved total number of arms) and evolution in both DNA sequence and copy number of the pericentromeric alpha-satellite I repetitive DNA sequence. Here, a probe representing the sheep alpha-satellite I sequence was isolated and hybridized to genomic DNA digests and metaphase chromosomes from various Bovidae species. The probe was highly homologous to the centromeric sequence in all species in the tribe Caprini, including sheep (Ovis aries), goat (Capra hircus) and the aoudad or Barbary sheep (Amnotragus lervia), but showed no detectable hybridization to the alpha-satellite I sequence present in the tribe Bovini and at most very weak to species in the tribes Hippotragini, Alcelaphini or Aepycerotini. The sex chromosomes of sheep, goat and aoudad did not contain detectable alpha-satellite I sequence; in sheep, one of the three metacentric autosomal chromosomes does not carry the sequence, while in aoudad, it is essentially absent in three large autosomal pairs as well as the large metacentric chromosome pair. The satellite probes can be used as robust chromosome and karyotype markers of evolution among tribes and increase the resolution of the evolutionary tree at the base of the Artiodactyla.

Centromere emergence in evolution

Evolutionary centromere repositioning is a paradox we have recently discovered while studying the conservation of the phylogenetic chromosome IX in primates. Two explanations were proposed: a conservative hypothesis assuming sequential pericentric inversions, and a more challenging assumption involving centromere emergence during evolution. The complex evolutionary history showed by chromosome IX did not allow us to clearly distinguish between these two hypotheses. Here we report comparative studies of chromosome X in two lemur species: the black lemur and the ringtailed lemur. The X chromosome is telocentric in the black lemur and almost metacentric in the ringtailed lemur. The marker order along these chromosomes, however, was found to be perfectly colinear with humans. Our data unequivocally point to centromere emergence as the most likely explanation of centromere repositioning.

Genetic differences between humans and great apes

The remarkable similarity among the genomes of humans and the African great apes could warrant their classification together as a single genus. However, whereas there are many similarities in the biology, life history, and behavior of humans and great apes, there are also many striking differences that need to be explained. The complete sequencing of the human genome creates an opportunity to ask which genes are involved in those differences. A logical approach would be to use the chimpanzee genome for comparison and the other great ape genomes for confirmation. Until such a great ape genome project can become reality, the next best approach must be educated guesses of where the genetic differences may lie and a careful analysis of differences that we do know about. Our group recently discovered a human-specific inactivating mutation in the CMP-sialic acid hydroxylase gene, which results in the loss of expression of a common mammalian cell-surface sugar throughout all cells in the human body. We are currently investigating the implications of this difference for a variety of issues relevant to humans, ranging from pathogen susceptibility to brain development. Evaluating the uniqueness of this finding has also led us to explore the existing literature on the broader issue of genetic differences between humans and great apes. The aim of this brief review is to consider a listing of currently known genetic differences between humans and great apes and to suggest avenues for future research. The differences reported between human and great ape genomes include cytogenetic differences, differences in the type and number of repetitive genomic DNA and transposable elements, abundance and distribution of endogenous retroviruses, the presence and extent of allelic polymorphisms, specific gene inactivation events, gene sequence differences, gene duplications, single nucleotide polymorphisms, gene expression differences, and messenger RNA splicing variations. Evaluation of the reported findings in all these categories indicates that the CMP-sialic hydroxylase mutation is the only one that has so far been shown to result in a global biochemical and structural difference between humans and great apes. Several of the other known genetic dissimilarities deserve more exploration at the functional level. Among the areas of focus for the future should be genes affecting development, mental maturation, reproductive biology, and other aspects of life history. The approaches taken should include both going from the genome up to the adaptive potential of the organisms and going from novel adaptive regimes down to the relevant repercussions in the genome. Also, as much as we desire a simple genetic explanation for the human phenomenon, it is much more probable that our evolution occurred in multiple genetic steps, many of which must have left detectable footprints in our genomes. Ultimately, we need to know the exact number of genetic steps, the order in which they occurred, and the temporal, spatial, environmental, and cultural contexts that determined their impact on human evolution.

Molecular structure and evolution of DNA sequences located at the alpha satellite boundary of chromosome 20

We have isolated and characterised one PAC clone (dJ233C1) containing a linkage between alphoid and non-alphoid DNA. The non-alphoid DNA was found to map at the pericentromeric region of chromosome 20, both on p and q sides, and to contain homologies with one contig (ctg176, Sanger Centre), also located in the same chromosome region. At variance with the chromosome specificity shown by the majority of non-alphoid DNA, a subset of alphoid repeats derived from the PAC yielded FISH hybridisation signals located at the centromeric region of several human chromosomes, belonging to three different suprachromosomal families. The evolutionary conservation of this boundary region was investigated by comparative FISH experiments on chromosomes from great apes. The non-alphoid DNA was found to have undergone events of expansion and transposition to different pericentromeric regions of great apes chromosomes. Alphoid sequences revealed a very wide distribution of FISH signals in the great apes. The pattern was substantially discordant with the data available in the literature, which is essentially derived from the central alphoid subset. These results add further support to the emerging opinion that the pericentromeric regions are high plastics, and that the alpha satellite junctions do not share the evolutionary history with the main subsets.

Mosaic telomeric (2;14) association in a child with motor delay

In a 6-year-old girl referred because of mild motor delay and hyperextensible joints, chromosome analysis disclosed a derivative chromosome consisting of end-to-end fusion of chromosomes 2 and 14. Two cell lines existed in which this telomere association was present, one with a 45,XX,tas(2;14)(q37;p11) karyotype and one with a 45,XX,tas(2;14) (q37;q32) karyotype. The cell line with the telomeric fusion of 2q and 14p was present in 90% of the cells; a telomeric fusion of 2q and 14q was seen in the remaining 10% of the cells. In both association complexes, only the centromere of chromosome 14 was active. Fluorescence in situ hybridization with telomere and subtelomere probes disclosed no deletion of chromosomal material. Microsatellite analysis showed that the patient had a normal biparental contribution of chromosomes 14.

Non-random trisomies of chromosomes 5, 8 and 12 in the prolactinoma sub-type of pituitary adenomas: conventional cytogenetics and interphase FISH study

Specimens from 53 pituitary adenomas (PAs), including 17 NFPA, 16 PRL-, 9 ACTH-, 9 GH- and 2 TSH-secreting tumors, underwent cytogenetic analysis by the direct and short-term culture methods. Only 8 tumors (15%) appeared to have an abnormal karyotype. To increase the resolution of cytogenetic analysis, direct preparations from 31 PAs were investigated by interphase FISH with probes specific for chromosomes 5, 8, 12 and X, for which gain in pituitary tumors has been reported. Of these 31 PAs, 17 (54.8%) had an abnormal dosage of one or more of the 4 chromosomes tested. Separate or combined trisomies of chromosomes 5, 8 and 12 were found in 10/10 prolactinomas and in 4/9 NFPA, whereas the combined loss of chromosomes 5 and 8 was observed in 1/6 ACTH- and 1/6 GH-secreting PAs. Present and earlier data on 23 PAs showed that tumors with the highest frequency of abnormal karyotypes revealed by cytogenetics and/or interphase FISH were PRL (78%), followed by NFPA (26%) and GH (18%). Recurrent structural rearrangements affecting chromosomes 1, 3 and 12 were also identified in prolactinomas, which therefore appear to be the only pituitary adenoma sub-type with a defined trend of tumor-specific chromosomal changes. Cytogenetic and FISH analyses of different pituitary tumor sub-types indicate that they may harbour genetically distinct lesions.

MFASAT: a new alphoid DNA sequence isolated from Macaca fascicularis (Cercopithecidae, Primates)

A new highly repeated DNA fragment isolated from Macaca fascicularis (MFASAT) is described. Our findings obtained by sequencing, Southern blot analysis, and fluorescent in situ hybridization (FISH) on metaphasic chromosomes strongly suggest that MFASAT can be considered as a member of the alphoid DNA family characteristic of Old World monkeys. The chromosomal localization of MFASAT, obtained by FISH, showed that this alphoid DNA is present in the peri-centromeric area of all the chromosomes. MFASAT showed a high degree of conservation when compared, by sequence alignment, to other Macaca species and Papio papio as expected for species with considerable genome conservation. A low degree of homology has been found comparing M. fascicularis alphoid DNA with a more distantly related Cercopithecidae species such as Cercopithecus aethiops.

Centromere repositioning

Primate pericentromeric regions recently have been shown to exhibit extraordinary evolutionary plasticity. In this paper we report an additional peculiar feature of these regions that we discovered while analyzing, by FISH, the evolutionary conservation of primate phylogenetic chromosome IX. If the position of the centromere is not taken into account, a relatively small number of rearrangements must be invoked to account for interspecific differences. Conversely, if the centromere is included, a paradox emerges: The position of the centromere seems to have undergone, in some species, an evolutionary history independent from the surrounding markers. A significant number of additional rearrangements must be proposed to reconcile the order of the markers with centromere position. Alternatively, the evolutionary emergence of neocentromeres can be postulated.

Characterization of constitutive heterochromatin in Cebus apella (Cebidae, Primates) and Pan troglodytes (Hominidae, Primates): comparison to human chromosomes

Using G bands, some homologies between the chromosomes of Cebus apella (CAP) and human chromosomes are difficult to establish. To solve this problem, we analyzed these homologies by fluorescence in situ hybridization using human whole chromosome probes (ZOO-FISH). The results indicated that 1) the human probe for chromosome 2 partially hybridizes with CAP chromosomes 13 and 5, 2) the human probe for chromosome 3 partially hybridizes with CAP chromosomes 18 and 20, 3) the human probe for chromosome 9 partially hybridizes with CAP chromosome 19, and 4) the human probe for chromosome 14 hybridizes with the p-terminal and q-terminal regions of CAP chromosome 6. However, none of the human probes employed hybridized with the heterochromatic regions of CAP chromosomes. For this reason, we characterized the heterochromatic regions of CAP chromosomes and of the chromosomes of Pan troglodytes (PTR), to allow comparison between CAP, PTR, and human chromosomes using in situ digestion of fixed chromosomes with the restriction enzymes AluI, HaeIII, and RsaI and by fluorescent staining with DA/DAPI. The results show that 1) centromeric heterochromatin is heterogeneous in the three species studied and 2) noncentromeric heterochromatin is homogeneous within each of the three species, but is different for each species. Thus, centromeric heterochromatin undergoes a higher degree of variability than noncentromeric heterochromatin.

Conservation of human gamma-X centromeric satellite DNA among primates with an autosomal localization in certain Old World monkeys

Gamma-X satellite DNA is a 220-bp tandemly arranged repetitive DNA with specificity for the centromeric region of the human X chromosome. The conservation of this human X centromeric satellite DNA sequence in primate species was evaluated by comparative fluorescence in-situ hybridization to metaphase chromosome preparations of the great apes and three Old World monkeys. Homologous gamma-X DNA were detected at centromeric locations in all six primate species. For the great apes, gamma-X was exclusively localized to the centromeric regions of the X chromosomes. Among the Old World monkeys studied, only the golden monkey exhibited localization to the X chromosome. In the black-and-white colobus and the pig-tailed macaque, human gamma-X sequences were localized to the pericentromeric regions of autosomes 1 and 4, respectively.

Genetic and physical analyses of the centromeric and pericentromeric regions of human chromosome 5: recombination across 5cen

Human centromeres are poorly understood at both the genetic and the physical level. In this paper, we have been able to distinguish the alphoid centromeric sequences of chromosome 5 from those of chromosome 19. This result was obtained by pulsed-field gel electrophoresis after cutting genomic DNA with restriction endonucleases NcoI (chromosome 5) and BamHI (chromosome 19). We could thus define a highly polymorphic marker, representing length variations of the D5Z1 domain located at the q arm boundary of the chromosome 5 centromere. The centromeric region of chromosome 5 was then analyzed in full detail. We established an approximately 4.6-Mb physical map of the whole region with five rare-cutting enzymes by using nonchimeric YACs, two of which were shown to contain the very ends of 5cen on both sides. The p-arm side of 5cen was shown to contain an alphoid subset (D5Z12) different from those described thus far. Two genes and several putative cDNAs could be precisely located close to the centromere. Several L1 elements were shown to be present within alpha satellites at the boundary between alphoid and nonalphoid sequences on both sides of 5cen. They were used to define STSs that could serve as physical anchor points at the junction of 5cen with the p and q arms. Some STSs were placed on a radiation hybrid map. One was polymorphic and could therefore be used as a second centromeric genetic marker at the p arm boundary of 5cen. We could thus estimate recombination rates within and around the centromeric region of chromosome 5. Recombination is highly reduced within 5cen, with zero recombinants in 58 meioses being detected between the two markers located at the two extremities of the centromere. In its immediate vicinity, 5cen indeed exerts a direct negative effect on meiotic recombination within the proximal chromosomal DNA. This effect is, however, less important than expected and is polarized, as different rates are observed on both arms if one compares the 0 cM/Mb of the p proximal first 5.5 Mb and the 0.64 cM/Mb of the q proximal first 5 Mb to the sex-average 1.02 cM/Mb found throughout the entire chromosome 5. Rates then become close to the average when one goes further within the arms. Finally, most recombinants (21/22), irrespective of the arm, are of female origin, thus showing that recombination around 5cen is essentially occurring in the female lineage.

Numerical abnormalities of chromosomes 1 and 10 in endometrial adenocarcinoma: fluorescence in situ hybridization analysis of 23 archival paraffin-embedded samples

Retrospective analysis of chromosomal changes in endometrial carcinoma was performed by fluorescence in situ hybridization on free nuclei isolated from formalin-fixed paraffin-embedded tissue. We examined 23 archival samples for numerical aberrations of chromosomes 1 and 10 with the use of specific DNA probes for the pericentromeric and centromeric regions of these two chromosomes. Numerical aberrations of chromosomes 1 and 10 were detected in 39% of the case analyzed, and the frequency of trisomy 10 tended to increase as the histological grade worsened. Our findings confirm the association of cytogenetic anomalies involving chromosomes 1 and 10 with endometrial carcinoma, as reported by other studies, and suggest that changes in centromere 10 copy number may correlate with the degree of tumor differentiation.

Characterization of fetal haematopoietic progenitors circulating in maternal blood of seven aneuploid pregnancies

A retrospective study was carried out in order to investigate the phenotype of fetal haematopoietic progenitors circulating in the maternal blood of seven aneuploid pregnancies. Five of the blood samples were taken during pregnancies affected by various fetal aneuploidies, while the other two were collected after therapeutic abortion due to prenatal cytogenetic diagnosis of trisomies 21 and 18. Haematopoietic progenitor cells, isolated by labelling the erythropoietin receptors with the biotinylated ligand before magnetic sorting and/or fibronectin cell adhesion assay, were cultured in a suitable semisolid medium. Single- or dual-colour fluorescence in situ hybridization (FISH) was utilized to identify and enumerate fetal cells amplified in culture. Fetal trisomies were confirmed in the FISH analysis with chromosome-specific probes in all the cases analysed. The fetal purity rate ranged from 16 to 26 per cent. Haematopoietic progenitors of fetal origin were found to include CFU-E, CFU-GEMM, and possibly also M-BFU-E. Interestingly, a more immature progenitor with high self-renewal capacity (CFU-blast cell) isolated by fibronectin sorting was shown to have a relatively high frequency in one case of Down syndrome. In general, the results of this study demonstrate the feasibility of diagnosing the major fetal chromosomopathies by culturing fetal cells taken from maternal blood. Furthermore, our initial data on the sequential sorting for fibronectin and erythropoietin receptors lead us to believe that this approach may broaden the range of fetal haematopoietic progenitors retrievable from the maternal circulation.

Maternal uniparental disomy for chromosome 22 in a child with generalized mosaicism for trisomy 22

We report on a case of generalized mosaicism for trisomy 22. At chorionic villus sampling (CVS) in the 37th week of pregnancy, a 47,XX,+22 karyotype was detected in all cells. The indication for CVS was severe unexplained symmetrical intrauterine growth retardation (IUGR) and a ventricular septal defect (VSD) was noted. In cultured cells from amniotic fluid taken simultaneously, only two out of ten clones were trisomic. At term, a growth-retarded girl with mild dysmorphic features was born. Lymphocytes showed a normal 46,XX[50] karyotype; both chromosomes 22 were maternal in origin (maternal uniparental disomy). Investigation of the placenta post-delivery using fluorescence in situ hybridization showed a low presence of trisomy 22 cells in only one out of 14 biopsies. In cultured fibroblasts of skin tissue, a mosaic 47,XX,+22[7]/46,XX[25] was observed. Clinical follow-up is given up to 19 months.

Search for neuroblastoma loci: characterization of tumor cell lines that could facilitate their positional cloning

Specific chromosomal aberrations might indicate the position of genes responsible for a particular disease. Neuroblastoma is characterized by frequent deletions and/or rearrangements of the subtelomeric 1p region which, accordingly, is believed to host one or more oncosuppressor gene(s) directly or indirectly involved in the development of this and other tumors. Identification of these genes could be facilitated if cell lines with well characterized interstitial deletions or reciprocal translocations could be available for application of positional cloning strategies. In the present report we present additional and novel molecular data on three well established neuroblastoma cell lines (NLF, NMB and NGP). In one of these we have identified two sites that might be good candidates for hosting oncosuppressor genes; one of these is flanked by the D1S47 and ENO1 loci while the other is distal to the A12M2 locus.

Molecular-cytogenetic characterization of a higher plant centromere/kinetochore complex

The centromeric region of a telocentric field bean chromosome that resulted from centric fission of the metacentric satellite chromosome was microdissected. The DNA of this region was amplified and biotinylated by degenerate oligonucleotide-primed polymerase chain reaction (DOP-PCR)/linker-adapter PCR. After fluorescence in situ hybridization (FISH) the entire chromosome complement of Vicia faba was labelled by these probes except for the nucleolus organizing region (NOR) and the interstitial heterochromatin, the chromosomes of V. sativa and V. narbonensis were only slightly labelled by the same probes. Dense uniform labelling was also observed when a probe amplified from a clearly delimited microdissected centromeric region of a mutant of Tradescantia paludosa was hybridized to T. paludosa chromosomes. Even after six cycles of subtractive hybridization between DNA fragments amplified from centromeric and acentric regions no sequences specifically located at the field bean centromeres were found among the remaining DNA. A mouse antiserum was produced which detected nuclear proteins of 33 kDa and 68 kDa; these were predominantly located at V. faba kinetochores during mitotic metaphase. DNA amplified from the chromatin fraction adsorbed by this serum out of the sonicated total mitotic chromatin also did not cause specific labelling of primary constrictions. From these results we conclude: (1) either centromere-specific DNA sequences are not very conserved among higher plants and are - at least in species with large genomes - intermingled with complex dispersed repetitive sequences that prevent the purification of the former, or (2) (some of) the dispersed repeats themselves specify the primary constrictions by stereophysical parameters rather than by their base sequence.