INTRODUCTION

Rhoeo spathacea (Swartz) Stearn. (= Rhoeodiscolor) has been a classical subject of cytologi-cal studies for about 70 years since the pioneeringworks of BELLING (1927), DARLINGTON (1929)and SAX (1931). It is widely believed that thepermanent structural heterozygosity of thisspecies (the lack of chromosome pairs and unusu-al mode of meiosis, with chromosome rings orchains at the first division) is conditioned by aseries of reciprocal translocations between thedistal segments of all chromosomes (for review:SATTERFIELD and MERTENS 1972). Most of thedata on karyotype structure in this species havebeen obtained from observations of convention-ally stained meiotic chromosomes (KATO 1930;SAX 1931, 1935; COLEMAN 1941; LIN and PAD-DOCK 1973, 1978; LIN 1979). Besides numerousnon-interstitially located heterochromatic bands(NATARAJAN and NATARAJAN 1972; STACK 1974;PETTENATTI 1987; GOLCZYK and JOACHIMIAK

1999), no other coding and non-coding domainshave been localized within the karyotype of thisinteresting species.

There are some reports on the localization ofnucleolar organizers (NORs) in Rhoeo, but only

from indirect observations in unrelated studies.DARLINGTON (1929), BHADURI (1942) and LIN

and PADDOCK (1973) found several interstitiallylocated, weakly stained regions within some con-ventionally stained meiotic/mitotic chromosomes,and interpreted them as secondary constrictions(“NORs”). However, the constrictions localizedby LIN and PADDOCK (1973) significantly differfrom those detected by BHADURI (1942) and DAR-LINGTON (1929). On the other hand, STACK

(1974), based on the reasonable generalizationthat nucleolus organizers in plants are usuallyassociated with constitutive heterochromatin, hasquestioned the insterstitial localization of NORsin this species and suggested the presence of onlytwo terminal NOR loci on two submetacentricchromosomes equipped with telomeric hete-rochromatin on the shorter arms.

Secondary constrictions may be viewed as astate of arrest of condensation of the regionsbearing the nucleoli at prophase, that is, they canbe treated as chromosomal markers of activeNORs within conventionally stained preparations(BUSH and SMETANA 1970). Although not allinterstitially located active NORs are marked bya secondary constriction, all true secondary con-strictions should be treated as active NOR sites.Transcriptionally inactive rDNA clusters, if pre-sent, do not form secondary constrictions, andsatellites are not observed within the chromo-

CARYOLOGIA Vol. 56, no. 1: 31-35, 2003

NORs in Rhoeo (Commelinaceae) revisited

HIERONIM GOLCZYK* and ANDRZEJ JOACHIMIAK

Department of Plant Cytology and Embryology, Jagiellonian University, ul. Grodzka 52, 31-044 Kraków, Poland.

Abstract - Nucleolar organizer regions were localized by silver staining in Rhoeospathacea (Swartz) Stearn. Our results strongly suggest that interstitially locatedlightly stained chromosome segments observed in conventional preparationsand interpreted by previous authors as secondary constrictions are depleted ofactive nucleolus organizing regions. There were no Ag-positive intercalary signalswithin chromosome arms where lightly stained chromosome segments wereobserved. Ag-NOR loci were found only at telomeric and centromeric chromo-some domains, and colocalized with heterochromatin bands.

Key words: Ag-staining, chromosomes, heterochromatin, karyotype, NORs,Rhoeo spathacea.

* Corresponding author: fax ++48-12-4228107, e-mail:h.golczyk@iphils.uj.edu.pl

somes bearing them (LENGEROVA and VYSKOT

2001). The intercalary localization of NORs inRhoeo suggested by the majority of previousauthors (l.c.) seems in disagreement with the lackof intercalary heterochromatin (STACK 1974; PET-TENATI 1987; GOLCZYK and JOACHIMIAK 1999),the apparent lack of interstitial CMA-positive sig-nals (JOACHIMIAK and GOLCZYK, unpbl. results)and also the absence of stable types of satellitedchromosomes in conventionally stained mitoticpreparations (GOLCZYK and JOACHIMIAK 1999).Thus it seems probable that the previouslydescribed “secondary constrictions” in Rhoeo areartifacts or else represent chromatin segments ofunknown structure but lacking active nucleolarorganizers. For the first time we use silver stain-ing (Ag-staining), a simple technique specificallydetecting functional nucleolus organizers (CHE-UNG et al. 1989) within the Rhoeo karyotype toresolve this problem.

MATERIALS AND METHODS

Cytological methodsRoots were taken from cuttings of five different R.

spathacea plants we previously analyzed by conven-

tional and C-banding methods (GOLCZYK andJOACHIMIAK 1999). Floral buds were obtained fromthe same donor plants. Root tips and anthers werefixed in AA (glacial acetic acid and absolute alcohol,1:3) for 2 h, stained in aceto-orcein (2% solution oforcein in 45% acetic acid) at room temperature andsquashed in a drop of 45% acetic acid with no heat-ing to minimize the risk of chromosome breaks. ForAg-staining the material was fixed overnight refriger-ated in FAA (50% alcohol, glacial acetic acid and37% formaldehyde, 18:1:1), squashed and stainedaccording to Trude Schwarzacher and KaryobiologyGroup (http://www.le.ac.uk/bl/phh4/methods/agn-or.htm). The twentyfive best mitotic metaphases andten best meiotic chains were selected and captured ina Lucia G system (Laboratory Imaging Ltd., CzechRepublic) or photographed. For C-banding the mate-rial was fixed in AA for 2 h and further treatedaccording to the schedule described by SCHWARZACH-ER et al. (1980).

Data processing and presentationImages were captured and processed using a CCD

camera and Lucia G. The chromosome lengths werecalculated as percentages of total karyotype length.Karyotype analysis was performed on 10 completemeiotic and 25 mitotic metaphases. Chromosometypes were identified and classified as previouslydescribed (GOLCZYK and JOACHIMIAK 1999).

32 GOLCZYK and JOACHIMIAK

Fig. 1 – Somatic Rhoeo metaphase chromosomes with clear-ly visible LSSs. Acetoorceine staining. Bar 10µm.

Fig. 2 – Karyotype of Rhoeo spathacea. Chromosomesordered according to their relative position within meioticrings/chains. Black – chromosome regions with the highestfrequency of LSSs (above 0.4).

Table 1 – Rhoeo spathacea, acetoorceine stained chromosome types (1-12) showing different number of LSSs within five chro-mosome regions (ChR) in mitosis (a) and meiosis (b).

1 2 3 4 5 6 7 8 9 10 11 12

ChR a b a b a b a b a b a b a b a b a b a b a b a b1 6 4 1 1 1 1 3 1 5 1 1 0 1 0 1 3 1 0 0 0 1 0 0 02 11 2 13 3 12 2 17 1 18 4 10 2 9 1 11 0 5 0 3 0 3 0 11 23 1 0 10 0 7 0 1 0 16 0 2 0 0 0 1 0 0 0 0 0 0 0 2 04 11 0 0 0 1 0 11 2 2 0 1 0 7 1 1 0 0 0 8 0 2 0 1 05 1 2 0 0 0 0 1 0 1 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0

Total 30 8 24 4 21 3 33 4 42 5 14 2 18 2 14 3 6 0 11 0 7 0 14 2

NORs IN RHEO REVISITED 33

Fig. 3 – Chromosomes (A-D) and nuclei (E-H) of Rhoeo spathacea. A-C, silver-stained mitotic metaphase plates with two (A),three (B), and five (C) distinct telomeric Ag-NOR signals. Note the large silver positive centromeric signal in B (arrowed) and oth-er centric signals appearing as tiny double dots. D, C-banded mitotic metaphase plate. E-G, interphase nuclei from the root meris-tem showing different numbers of silver-stained nucleoli. H, mitotic prophase nucleus with large, collective nucleolus. Bars 10µm.

The lightly stained chromosome segments (LSSs)we observed (see Introduction) were found in mostaceto-orcein stained metaphases. The LSS positionswithin each analyzed chromosome were calculated asthe distance from the end of the longer chromosomearm. Each chromosome type was divided into fiveregions of equal length, the first region beginning atthe end of the longer arm (Fig. 2, Table 1). The expect-ed vs. observed distribution of LSSs along the distin-guished regions of chromosomes was tested with thechi-square test for goodness of fit (SOKAL and ROHLF

1969). The expected distribution was an even one,that is, an equal chance that an LSS would appear ina given chromosome region.

RESULTS

We observed LSSs (Fig. 1) within all conven-tionally stained mitotic metaphases and in sixmeiotic rings/chains. Within meiotic chromo-somes these structures were not easy for distinc-tion and they were less frequently observed(Table 1); thus mitotic metaphases seem moresuitable for analysis of their distribution. MitoticLSSs were clearly unstable as to their numberand localization within chromosomes. Their chro-mosome positions are highly variable, so finallywe distinguished only some chromosome regionscharacterized by relatively high (above 0.4) LSSfrequency (Fig. 2). The number of chromosomeswith LSSs in a given metaphase varied from 3 to11, and the number of these structures within aparticular chromosome was from 1 to 4. All chro-mosome variants with “secondary constrictions”described by previous authors (DARLINGTON

1929; BHADURI 1942; LIN and PADDOCK 1973)were observed in our study, but some of them ata very low frequency. The observed distributionof LSSs differed significantly from the expectedeven distribution (chi-square test = 176, s.s. = 4,p < 0.0001). The pattern of departure from theexpected distribution points to the centraldomains of the longer chromosome arms, wherean excess of LSSs is manifested.

All the Ag-positive signals were located atchromosome termini and/or within centromericclusters of chromatin (Figs. 3A, B, C). There wereno positive interstitially located signals withinarms, where LSSs were frequently observed inorcein-stained preparations. In mitotic chromo-somes large signals were located mainly at telom-eres, whereas centric signals were weak anddetected chiefly in the form of tiny double dots

(Figs. 3A, B, C). The number of potentially activenucleolus organizers in Rhoeo (identified withindifferent chromosome positions) seems relative-ly high; this may be additionally supported bythe observations of interphase nuclei (Figs. 3E, F,G). In the majority of root-tip interphase cellsthe number of silver-stained nucleoli was 1-6,markedly exceeding STACK’s (1974) estimates butin accordance with BHADURI (1942), who statedthat the maximum number of separate nucleoliobserved in Rhoeo root meristem is 6. In fact,the strong tendency to nucleolar fusion in Rhoeo(Figs. 3G, H) obscures the situation; nucleolicould be even more abundant (BHADURI 1942).

Centromerically located Ag-NOR signals colo-calized with previously described (GOLCZYK andJOACHIMIAK 1999) centric blocks of heterochro-matin, but in C-banded mitotic chromosomessuch heterochromatin clusters are markedly larg-er than the majority of centric Ag-positive signalsdetected here (Fig. 3D). Thus the possibility thatheterochromatin in Rhoeo is agyrophilic per secan be ruled out. Terminal Ag-NOR signals couldalso be traced to the heterochromatin becausefive terminally located segments of heterochro-matin have been carefully localized within theRhoeo karyotype (GOLCZYK and JOACHIMIAK

1999). Unfortunately, exact identification ofNOR-bearing Rhoeo chromosomes by morphol-ogy is very difficult in our silver-stained prepara-tions, and further studies are needed.

DISCUSSION

It is broadly accepted that Ag-NOR-positivechromosome signals correspond to the locationsof active rRNA genes (CHEUNG et al. 1989), andthat the size and intensity of the Ag-stained areasare related to their transcriptional activity(MILLER et al. 1976; HOFGÄRTNER et al. 1979).The observed exclusively proximal and distallocalization of Ag-stained regions strongly sug-gests that interstitially located LSSs are not truesecondary constrictions (structures organizingthe nucleolus). These lightly stained bandsobserved by us and by previous authors (DAR-LINGTON 1929; BHADURI 1942; LIN and PADDOCK

1973) proved surprisingly unstable as to theirnumber and localization within differentmetaphase plates. The nature of these bandsremains unknown; their distribution differs sig-nificantly from the expected random distribu-

34 GOLCZYK and JOACHIMIAK

tion (see Results). Even if the LSSs are breakscaused by the method used, their non-randomdistribution and abundant occurrence within def-inite chromosome segments suggest increasedsusceptibility of these segments to acidic treat-ment.

Acknowledgments – We are indebted to Dr. J.Mitka for help and valuable discussion.

REFERENCES

BELLING J., 1927 – The attachement of chromosomesat the reduction division in flowering plants. Jour-nal of Genetics, 18: 177-205.

BHADURI P.N., 1942 – Cytological analysis of struc-tural hybridity in Rhoeo discolor Hance. Journalof Genetics, 44: 73-85.

BUSH H. and SMETANA K., 1970 – Nucleolar orga-nizers, nucleolar size, shape and number. In: H.Bush H and K. Smetana (Eds.), “The nucleo-lus”, pp. 115-159. Academic Press, New Yorkand London.

CHEUNG S.W., SUN L. and FEATHERSTONE T., 1989– Visualization of NORs in relation to the precisechromosomal localization of ribosomal RNAgenes. Cytogenetics and Cell Genetics, 50: 93-97.

COLEMAN L.C., 1941 – The relation of chromocentersto the differential segments in Rhoeo discolorHance. American Journal of Botany, 28: 742-8.

DARLINGTON C.D., 1929 – Chromosome behaviourand structural hybridity in the Tradescantiae.Journal of Genetics, 21: 207-286.

GOLCZYK H. and JOACHIMIAK A., 1999 – Karyotypestructure and interphase chromatin organizationin Rhoeo spathacea (Sw.) Stearn (Commeli-naceae). Acta Biologica Cracoviensia SeriesBotanica, 41: 143-150.

HOFGÄRTNER F.J., KRONE W. and JAIN K., 1979 –Correlated inhibition of ribosomal RNA synthesisand silver staining by actinomycin D. HumanGenetics, 47: 329-333.

KATO K., 1930 – Cytological studies of pollen moth-er cells of Rhoeo discolor Hance, with special ref-erence to the question of the mode of syndesis.

Memoirs of the Collegium of Science of KyotoImperial University Series B, 5: 139-161.

LENGEROVA M. and VYSKOT B., 2001 – Sex chro-matin and nucleolar analyses in Rumex acetosa L.Protoplasma, 217: 147-153.

LIN Y.J., 1979 – Chromosome distribution and cate-nation in Rhoeo spathacea var. concolor. Chro-mosoma, 71: 109-127.

LIN Y.J. and PADDOCK E.F., 1973 – Ring-positionand frequency of adjacent distribution of meioticchromosomes in Rhoeo spathacea. AmericanJournal of Botany, 60: 685-690.

–, 1978 – Identification of complexes in a sponta-neous triploid Rhoeo spathacea. Chromosoma,67: 97-108.

MILLER O.J., MILLER D.A., DEV V.G., TANTRAVAHIR. and CROCE C.M., 1976 – Expression of humanand suppression of mouse nucleolus organizeractivity in mouse-human somatic cell hybrids.Proceedings of the National Academy of Sci-ences U.S.A., 73: 4531-4553.

NATARAJAN A.T. and NATARAJAN S., 1972 – Theheterochromatin of Rhoeo discolor. Hereditas,72: 323-330.

PETTENATI M.J., 1987 – Giemsa C - banding ofRhoeo (Commelinaceae). Genetica, 74: 219-224.

SATTERFIELD S.K. and MERTENS T.R., 1972 – Rhoeospathacea: A tool for teaching meiosis and mitosis.Journal of Heredity, 63: 375-378.

SAX K., 1931 – Chromosome ring formation inRhoeo discolor. Cytologia, 3: 36-53.

–, 1935 – Chromosome structure in the meiotic chro-mosomes of Rhoeo discolor Hance. Journal ofArnold Arboretum, 16: 216-224.

SCHWARZACHER T., AMBROS P. and SCHWEIZER D.,1980 – Application of Giemsa banding to orchidkaryotype analysis. Plant Systematics and Evo-lution, 134: 293-297.

SOKAL R.R. and ROHLF F.J., 1969 – Biometry. Theprinciples and practice of statistics in biologicalresearch. Freeman W.H. and Co., San Francisco.

STACK S.M., 1974 – Differential giemsa staining ofkinetochores and nucleolus organizer heterochro-matin in mitotic chromosomes of higher plants.Chromosoma, 47: 361-378.

Received June 15, 2002; accepted August 20, 2002

NORs IN RHEO REVISITED 35