Cytogenetic study of the genus Anchusa L. (Boraginaceae) in Iran

Abstract

Anchusa L., with four species, is one of the weedy species of Boraginaceae in Iran. These species are distributed in different habitats. Species delimitation is mainly based on corolla lobes, corolla tube condition and fruit shape. In this study chromosome number and meiotic behavior of 10 populations of Anchusa taxa have been considered. Due to the large number of studied populations of A. italica we have considered the intra-specific variation in this species. Statistical multivariate methods were done by use of PAST software ver. 5. 2012. Although all populations showed the chromosome number 2n = 4x = 32, population of A. arvensis subsp. orientalis showed the presence of n = 8 (2n = 2x = 16). This is the first meiotic study for Anchusa. Variation in meiosis stages such as synozytic knot was observed. Meiotic irregularities such as cytomyxis, stickiness, anaphase bridge, laggard chromosomes, micronucleus, triad and unreduced pollen grains were observed.

Keywords:

• Anchusa
• Iran
• chromosome number
• meiosis

Introduction

Anchusa L. (Boraginaceae) is a large genus with about 170 annual and perennial species in temperate and sub-temperate zone of the Old World (Akcin et al. 2010). Southern parts of Balkan Peninsula are the main center of diversification of this genus (Selvi and Bigazzi 2003). Anchusa species are distributed in Europe, north and south of Africa and West of Asia. These weedy plants (annual, biennial or perennial species) are covered with robust hairs and rough nodes (Judd et al. 1999). Leaves are ovate or lanceolate and sessile on top of the stem. Blue or white flowers aggregate in cymes. The fruit is composed of four nutlets with wrinkled surface.

Anchusa is composed of four species and five taxa (A. italica has two varieties: italica and kurdica) in Iran, which is distributed in the north, northwest, center, northeast and south of Iran. A. italica var. italica is more frequent than the others (Khatamsaz 2002). Anchusa species have ornamental, medicinal and edible uses in different parts of Europe and Asia. In Iran these species are used as a substitution for the medicinal Borago (Zargari 1989). Moreover, Anchusa roots have a special resinous component as a coloring agent. Some species are also weeds of different cultivations (Judd et al. 1999). Riedl (1967) and Khatamsaz (2002) and many other researchers mentioned different classifications at infra-generic and infra-specifics levels in Anchusa (Guşuleac 1927; Chater 1972; Greuter et al. 1984; Selvi and Bigazzi 1998).

No research on gametophyte content has been reported on Anchusa species. Only Ghaffari (1996) reported n = 16 for A. italica. Several cytogenetic studies have been performed on Anchusa from different parts of the world (Luque 1980; D’Amato and Trojani 1985; D’Amato 1989; Diaz lifante et al. 1992; Constantinidis and Kamari 1995; Markova and Goranova 1995; Ghaffari 1996; Lövkvist and Hultgård 1999), but such studies are not available for the Anchusa species growing in Iran. In this study accessions of Anchusa are collected from different locations of Iran and chromosomal behavior and gametophyte content of Anchusa species are reported for the first time.

Methods and materials

Plant material

Meiotic studies were performed in 10 populations of four Anchusa species and varieties of Iran (Table 1). The studied species are: (1) A. italica Retz. var. italica (six populations); (2) A. italica var. kurdica Gusuleac. (two populations); (3) A. strigosa Labill. (one population); and (4) A. arvensis subsp. orientalis (L.) Nordh. (one population). The voucher specimens are deposited in Herbarium of Alzahra University (AUH).

Table 1. List of Anchusa species studied.

Number of population	Species	Locality	Voucher no	Collector and Date
1	A. italica var. italica	Tehran, Vanak, 1300 m	AUH 13910	Nasrollahi, May 2010
2	A. italica var. italica	Tehran, Sadat Abad, 1250 m	AUH 13913	Nasrollahi, May 2010
3	A. italica var. italica	Hamedan, Varfejin, Emamzadeh Kouh, 1741 m	AUH 13909	Keshavarzi, June 2010
4	A. italica var. italica	Tehran, Darband, 1700 m	AUH 13914	Nasrollahi, April 2010
5	A. italica var. italica	Tehran, Shahrestanak, 2190 m	AUH 13911	Nasrollahi, May 2010
6	A. italica var. italica	Tehran, Dizin, 2650 m	AUH 13912	Nasrollahi, April 2010
7	A. italica var. kurdica	Kurdistan, Kamyaran, Yaghvasi village, 1464 m	AUH 13920	Nasrollahi, July 2010
8	A. italica var. kurdica	Kurdistan, Sanandaj, Kilak village, 1450 m	AUH 13921	Nasrollahi, July 2010
	A. arvensis subsp. orientalis	Kermanshah to Sanandaj, 1200 m	AUH 13908	Abassian, June 2010
	A. strigosa	Qazvin, Mohamadiyeh 15 km to Qazvin, 1300 m	AUH 13905	Nasrollahi, April 2010

Cytological studies

Young flower buds were collected from 10 randomly selected plants of each population, fixed in glacial acetic acid: ethanol (1:3) for 24 h, and then washed and preserved in 70% ethanol at 4°C until use, following Sheidai et al. (2003). Cytological preparations were done using squash technique and 2% aceto-orcein. Between 50 and 100 pollen mother cells (PMCs) were analyzed for chiasma frequency and distribution at diakinesis/metaphase stage, and 500 PMCs were analyzed for chromosome segregation during the anaphase and telophase stages.

Pollen stainability as a measure of fertility was determined by staining minimum 1000 pollen grains with 2% acetocarmine:50% glycerin (1:1) for about 30 min. Round complete pollens which were stained were taken as fertile, while incomplete, shrunken pollens with no stain were considered as sterile (Sheidai et al. 2003).

Statistical analyses

Grouping the populations of A. italica var. italica was performed using un-weighted paired group with arithmetic mean (UPGMA) as well as ordination based on principal component analysis (PCA) (Sheidai et al. 2002). For numerical analyses meiotic characteristics such as chiasmata frequency and distribution (terminal and intercalary chiasmata) and chromosome association (ring and rod bivalents) as well as univalents were used.

For grouping populations of four species of Anchusa, cluster analysis and PCA ordination were performed using relative meiotic characters (i.e. dividing the mean values of chiasmata frequency and bivalents by the chromosome number to obtain meiotic values per chromosome) unaffected by the chromosome number (Sheidai et al. 2002). Euclidean distance was used for cluster analysis. Statistical analyses were done using SPSS ver. 9 (1998). An ANOVA table is provided to show the significance variation (Table 2).

Table 2. ANOVA test of quantitative cytological features of the populations studied of Anchusa.

ANOVA
		Sum of squares	DF	Mean square	F	Significance
Total chiasmata	Between groups	792.662	3	264.221	4.406	0.007
	Within groups	4017.704	67	59.966
	Total	4810.366	70
Terminal chiasmata	Between groups	1273.304	3	424.435	14.284	0.000
	Within groups	1990.894	67	29.715
	Total	3264.197	70
Intercalary chiasmata	Between groups	957.136	3	319.045	34.829	0.000
	Within groups	613.737	67	9.160
	Total	1570.873	70
Univalent	Between groups	58.380	3	19.460	1.453	0.235
	Within groups	897.113	67	13.390
	Total	955.493	70
Rod bivalent	Between groups	347.726	3	115.909	29.255	0.000
	Within groups	265.457	67	3.962
	Total	613.183	70
Ring bivalent	Between groups	334.117	3	111.372	11.961	0.000
	Within groups	623.855	67	9.311
	Total	957.972	70
Quadrivalent	Between groups	5.989	3	1.996	7.778	0.000
	Within groups	17.195	67	0.257
	Total	23.183	70

Results

Studied Anchusa species showed 90% (A. italica var. italica, Hamedan population) to 99.7% (A. italica var. kurdica, Kilak village population) pollen fertility. Six populations of A. italica var. italica possessed n = 16 (2n = 4x = 32) chromosome number (Figure 1(A)), supporting an earlier report (Ghaffari 1996). More total and terminal chiasmata were observed in the Evin population and maximum ring bivalents were observed in the Darband population of A. italica var. italica (Table 3).

Table 3. Meiotic characters in Anchusa species studied.

Species	Locality	N	ROD	RB	I	Q	IX	TX	TOX	B	RODN	RBN	IN	QN	IXN	TXN	TOXN	PFP
A. italica var. italica	Hamedan	16	6.40	9.20	0.80	0.10	16.00	16.10	32.10	0.00	0.40	0.57	0.05	0.05	1.00	1.00	2.00	90.00%
A. italica var. italica	Evin	16	4.77	11.22	0.00	0.00	15.22	35.00	60.80	0.00	0.29	0.70	0.00	0.00	0.95	2.18	3.80	98.80%
A. italica var. italica	Deh Vanak	16	7.80	7.20	1.20	0.00	15.60	16.20	31.80	0.00	0.49	0.45	0.07	0.00	0.97	1.01	1.98	99.40%
A. italica var. italica	Shahrestanak	16	6.14	8.42	2.28	0.00	13.42	14.52	28.00	0.00	1.53	0.52	0.14	0.00	0.83	0.91	1.75	99.10%
A. italica var. italica	Dizin	16	4.50	4.50	14.00	0.25	8.75	8.00	16.75	0.00	0.28	0.28	0.87	0.01	0.54	0.50	1.04	97.20%
A. italica var. italica	Darband	16	2.25	11.75	4.00	0.00	13.25	21.75	35.00	0.00	0.14	0.73	0.25	0.00	0.82	1.35	2.18	99.30%
A. italica var. kurdica	Sanandaj, Kilak	16	2.00	12.50	1.75	0.75	15.00	22.25	37.25	0.00	0.12	0.78	0.10	0.04	0.93	1.39	2.32	99.70%
A. italica var. kurdica	Kamyaran	16	0.33	14.44	1.55	0.44	13.88	25.22	39.11	0.00	0.02	0.90	0.09	0.02	0.86	1.57	2.44	99.60%
A. strigosa	Qazvin	16	4.00	10.16	1.00	1.00	12.50	19.83	30.16	0.00	0.25	0.63	0.06	0.06	0.78	1.23	1.88	99.00%
A. arvensis	Sanandaj	8	0.33	13.88	0.00	0.00	3.44	27.00	30.44	0.00	0.04	1.73	0.00	0.00	0.43	3.37	3.80	99.20%

ROD = rod bivalent; RB = ring bivalent; I = univalent; Q = quadrivalent; IX = intercalary chiasmata; TX = terminal chiasmata; TOX = total chiasmata; B = number of B chromosomes; RODN = rod bivalent/cell; RBN = ring bivalent/cell; IN = univalent/cell; QN = quadrivalent/cell; IXN = intercalary chiasmata/bivalent; TXN = terminal chiasmata/bivalent; TOXN = total chiasmata/bivalent; PF = pollen fertility.

Two populations of A. italica var. kurdica studied possessed n = 16 (2n = 4x = 32; Figure 1(B), Table 1). Higher total and terminal chiasmata as well as ring bivalents were observed in the Kamyaran population (2.44, 1.57 and 0.90 respectively; Table 3) than in the Kilak village population. The only studied population of A. strigosa showed the presence of n = 16 (2n = 4x = 32), supporting the report of Diaz Lifante et al. (1992). The studied population of A. arvensis subsp. orientalis showed the presence of n = 8 (2n = 2x = 16; Figure 1(D)), supporting the earlier report of Luque (1980). Variation in chiasmata frequency has been reported in different accessions of some grasses, e.g. Aegilops, Lolium and Festuca (Rees and Dale 1974). Such variations could result in different kinds of recombinants, influencing the variability within natural populations in a possible adaptive manner (Rees and Dale 1974). Possible reasons for low pollen fertility in some accessions may be chromosome stickiness, laggard formation, triad and cytomixis.

Synezetic knot stage in meiosis-I prophase

The meiotic analysis of the studied Anchusa species showed variation in the prophase sub-stages of meiosis-I (Figure 2(A–C)). In the first sub-stage instead of the leptotene and zygotene stages a synezetic knot occurred, consisting of thin chromatin strands surrounding the nucleolus and eventually totally covering it (Figure 2(A)). Paired chromosomes unraveled from the knot and appeared as thick strands, ushering in the pachytene stage (Figure 2(B)), of which end-to-end attachment of chromosomes is a feature reported in taxa showing a synezetic knot stage (John et al. 1989; Sheidai and Inamdar 1991).

Figure 1. Representative pollen mother cells of Anchusa species showing gametic chromosome number in diakinesis-metaphase I. (A) Evin population of A. italica var. italica (n = 16); (B) Kilak population of A. italica var. kurdica (n = 16); (C) A. strigosa (n = 16); (D) A. arvensis subsp. orientalis (n = 8) (scale bar = 20 μm).

Figure 2. Representative pollen mother cells of Anchusa species showing prophase-I sub-stages of meiosis. (A) synezetic knot in Kilak population of A. italica var. kurdica; (B) pachytene stage in Evin population of A. italica var. italica; (C) diplotene stage in Evin population of A. italica var. italica (scale bar = 20 μm).

Figure 3. Representative meiotic cells in Anchusa species. (A) cytomictic cells of A. italica var. italica (Evin population); (B) extra chromosome in prophase I in A. arvensis subsp. orientalis; (C) micronucleus in A. italica var. kurdica (Kilak population); (D) anaphase bridge in A. italica var. italica (Darband population); (E) sticky chromosomes (telophase II) in A. italica var. kurdica (Kilak population); (F) micronucleus (telophase II) in A. strigosa; (G) tripolar cell; (H) metaphase I in A. italica var. italica (Vanak population) (Scale bar=20 μm).

Sticky chromosomes

Sticky chromosomes were observed from early stages of prophase and continued to the final stages of meiosis in most of the species studied. The number of chromosomes involved in stickiness varied from two to many, often forming a complete clumping of the chromosomes. Stickiness varied among different meiocytes and the species studied. Such a phenomenon has also been reported in Avena species (Baptista-Giacomelli et al. 2000; Sheidai et al. 2003). Genetic and environmental factors (Nirmala and Rao 1996), as well as genotype environment interactions (Baptista-Giacomelli et al. 2000), have been implicated as reasons for chromosome stickiness in different plant species, and this may be true for the studied Anchusa species as well.

Cytomixis

Chromatin/chromosome migration occurred in different directions from early prophase to telophase-II in almost all Anchusa species studied (Figure 3(A)). Several metaphase/diakinesis cells in these species possessed extra or missing chromosomes showing aneuploid condition (Figure 3(H)). Such aneuploid cells form aneuploid gametes/pollen grains, and several meiocytes with extra chromosomes were observed. The other result of cytomixis is the formation of cells with missing chromatin material, which leads to formation of abnormal tetrads, and infertile pollen grains.

Migration of chromatin material among the adjacent meiocytes occurs through cytoplasmic connections originating from the pre-existing system of plasmodesmata formed within the tissues of the anther. The plasmodesmata become completely obstructed by the deposition of callose, but in some cases they still persist during meiosis and increase in size, forming conspicuous inter-meiocytic connections or cytomictic channels that permit the transfer of chromosomes. Chromosome migration may also occur through cell wall dissolution among the neighboring meiocytes and forming syncyte (Falistocco et al. 1995).

Cytomixis is not considered to be of great evolutionary importance, but it may lead to production of aneuploid plants (Sheidai et al. 1993), or result in the production of unreduced gametes, as reported in several grass species (Falistocco et al. 1995).

Statistical analyses

UPGMA cluster analysis and PCA of the A. italica var. italica populations and of all species of Anchusa (performed separately) were done. In UPGMA Euclidean distance clustering dendrogram and PCA ordination (Figures 4 and 5), the n = 16 species (A. italica and A. strigosa) appear to be very similar and are placed very close to each other in a single cluster, while A. arvensis subsp. orientalis n = 8 appears to be more distant.

Figure 4. UPGMA cluster analysis of Anchusa species.

Figure 5. PCA ordination of Anchusa species.

The UPGMA dendrogram for the A. italica var. italica populations (Figure 6) shows two major clusters. The Dizin population is located a long way the first major cluster and forms a single cluster. Dizin was in higher geographic region with colder weather conditions, and most 2n pollen grains (about 1/5%) were observed in this population. These factors are suggested as the cause of this separation.

Figure 6. UPGMA cluster analysis of Anchusa italica var. italica populations.

PCA (Figure 5) of A. italica var. italica populations supports the cluster analysis showing separation of Dizin population from the others.

Unreduced pollen grain formation

The occurrence of large pollen grains (possibly 2n ones) were observed along normal ones in all studied Anchusa species (Figure 7). The large pollen grains comprised about 1% of pollen grains in these populations. The mean diameter of normal (reduced) pollen grains ranged from 19.97 μm to 71.12 μm, while the mean diameter of unreduced pollen grains ranged from 41.56 μm to 139.73 μm. 2n pollen production is indicated by presence of large pollen grains (Bretagnolle and Thomson 1995).

Figure 7. Potential unreduced (2n) pollen grain (large size pollen) in: (A) A. italica var. italica (Dizin population); (B) A. italica var. kurdica (Kamyaran population); (C) A. strigosa; (D) A. arvensis subsp. orientalis (scale bar = 60 μm).

Discussion

The present study showed the meiotic chromosome number of Anchusa species. Selvi et al. 2004 also pointed out the chromosome counts of some other taxa of Boraginaceae (Boragineae) on the phylogenetic tree to clarify the species relationship. As a final conclusion it should be mentioned that the chromosome number supported previous studies. The meiotic analysis revealed that in Anchusa species there were variations in the prophase sub-stages of meiosis-I, abnormal tetrads and also infertile pollen grains. Unreduced gamete formation is of evolutionary importance as it can lead to the production of plants with higher ploidy levels.

Unreduced gametes produce individuals with higher ploidy level by sexual polyploidization (Villeux 1985), as the major factor in formation of naturally occurring polyploids. Unreduced gametes are produced by different cytological mechanisms (Bretagnolle and Thomson 1995). The occurrence of multi-polar cells and cytomixis and irregularities in anaphase segregation of the chromosomes might be considered as the possible mechanisms of unreduced pollen grain formation in the studied Anchusa species. So far, this is the first report on the occurrence of unreduced pollen grains in Anchusa species.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This work was supported by Iran National Science Foundation [grant number 95s45031].