Printer Friendly

Low genetic diversity in Tillandsia recurvata (Bromeliaceae), the most ubiquitous epiphyte species of the semiarid and arid zones of North America.



Tillandsia recurvata (Bromeliaceae) is the most ubiquitous obligate epiphyte species (figure 1) of the arid and semiarid regions of North America (Benzing and Renfrow 1971, Soltis et al. 1987, Benzing 1990), whereas in the South American deserts, it is considered as a rare species (Pinto et al. 2006). In some areas, the high biomass load of this epiphyte eventually kills some branches or complete host-trees, for this reason the species has been considered a hemiparasite or structural parasite (Montana et al. 1997).

In the central Mexican semiarid region of the Zapotitlan Valley it was documented that the highest population density of Tillandsia recurvata is found on the legume trees of Prosopis laevigata, followed by the trees of Parkinsoniapraecox, and on the shrubs of Acacia bilimekii (Rzedowski 1978, Bernal et al. 2005). Occasionally, T. recurvata is found on Beaucarnea gracilis, and even on some columnar cacti as Cephalocereus columna-trajani and Neobuxbaumia tetetzp (Garcia-Suarez et al. 2003).

The differential population density of epiphytes on different host plant species has also been documented in other Tillandsia species. For example, the Mexican endemic bromeliad Tillandsia achyrostachys var. achyrostachys grows on 16 different host tree species, although the highest population density is found on Bursera copallfera (Gonzalez-Astorga et al. 2004). Another common example, in southern United States is T. usneoides which preferentially occurs on Quercus virginiana and Celtis spp (Callaway et al. 2002).


This differential population density observed on the different host species motivated us to explore whether it has any consequence in the population genetic attributes of Tillandsia recurvata. It is known that the genetic diversity in plants is influenced by their life history traits and breeding system (Loveless and Hamrick 1984, Bawa and Hadley 1991). Knowledge of pollination biology and seed dispersion for Tillandsia is fragmentary and scarce (Ramirez et al. 2008). T. recurvata has had some study done, and Soltis et al. (1987), based on its floral morphology, proposed that it is an autogamous species , with a wind seed dispersion mechanism. To date there have not been any field or experimental studies that this proposal. It should be noted that other Tillandsia species have been documented for self- compatibility and incompatibility, and vegetative reproduction is also present in this genus (Ramirez et al. 2008).

Only two population genetics studies are available for Tillandsia recurvata. The first one was carried out using 19 allozyme s loci in the northern Mexican populations, where a genetic diversity of zero was estimated (Soltis et al. 1987). In contrast, the other study carried out with a population from the Zapotitlan Valley of central Mexico, using only three microsatellite loci (Ramirez-Padilla, 2008), found an observed heterozygosis of 0.67 and an excess of heterozygotes as endogamy coefficient showed (FIS = -0.028). In addition, a genetic study available for the Mexican endemic T. achyrostachys found a low genetic diversity from 16 allozyme loci used (Gonzalez-Astorga et al. 2004).


In the present study, we evaluated the genetic diversity of Tillandsia recurvata on two host-tree species (Prosopis laevigata and Parkinsonia praecox). Based on population genetics theory predicts (Hedrick 2005), we expected to find higher genetic diversity values in the bromeliads populations established on P. laevigata, than those on P. praecox, because P. laevigata has lower population densities. We used microsatellite markers, short DNA sequences repeated in tandem with a typical length of 1-6 nucleotides scattered through the genome (Beebee and Rowe 2008). These molecular markers are appropriate for population genetic analyses because they detect high levels of plant polymorphism (Agarwal et al. 2008), can identify heterozygous individuals (Jarne and Lagoda 1996) and require small tissue samples (Jarne and Lagoda 1996; Agarwal et al. 2008).

Study site: This study was carried out in two localities located at the Zapotitlan valley in (14[degrees] 12' N, 92[degrees] 24' W). The bromeliad samples were collected from Prosopis. laevigata trees found in the alluvial terraces and from a close vegetation patch composed by Parkinsonia praecox. This valley is located in the southern portion of the Mexican State of Puebla, which is part of the Tehuacan-Cuicatlan biosphere reserve (figure 2). This reserve is in a semi-arid region of Mexico, and part of the Mexican xerophytic province (Rzedowski 1978). The Zapotitlan valley has an annual mean precipitation of 400 mm, with an average annual temperature of 21[degrees]C (Rzedowski 1978). In the mesquite bush the dominant trees are P. laevigata, P. praecox and some cacti species such as Mytillocactus geometrizans (figure 3), Pachycereus marginatus and Stenocereuspruinosus (Valiente-Banuet et al. 2000).


Methodology: In total, we collected 90 samples of Tillandsia recurvata, each one taken from a different tree. Samples were collected from 60 Prosopis laevigata trees (figure 4) and 30 Parkinsoniapraecox trees (figure 5). We collected bromeliad samples in all of the trees found in the study sites. We established as a T. recurvata population all clumps found among trees of each two species, and it is important to point out that this epiphyte was the only species found in all trees sampled. The populations of these two tree species are geographically separated by about 5 km so we considered the T. recurvata to be two different bromeliad populations.

Each bromeliad was wrapped individually in aluminum paper and stored in liquid nitrogen. The material was transported to the laboratory and maintained at -80 [degrees]C.

Genetic analyses: Total genomic DNA was isolated from 90 individuals of Tillandsia recurvata, using the DNeasy Plant Mini Kit (QIAGEN), according to the manufacturer's recommendations.


In order to find microsatellite regions in Tillandsia recurvata, we tested five primers (Table 1) previously designed for T. fasciculata and Guzmania monostachyya (Boneh et al. 2003). These primer pairs were assayed in individual PCR reactions containing 10 [iM of each primer, 100 uM of dNTPs mix, 2.0 mM of MgCl2, 40mM Tns-HCl pH 8.3, 40 mM of KCl, 100X BSA, 2 U Taq (Invitrogen) and 10-20 ng of total genomic DNA. The PCR cycles consisted of an initial denaturation at 94 [degrees]C for 2 min; 30 cycles consisting of denaturation at 94 [degrees]C for 10 s, annealing 51-58 [degrees]C for 10 s, and extension at 72 [degrees] C for 10 s; a final extension at 72 [degrees]C for 5 min. PCR products were run on 1.5% agarose gels with the inclusion of a 100 bp ladder size standard (Invitrogen). The forward primers were fluorescently labelled with fluorochromes HEX and FAM (Table 1). The electrophoretic reactions with fluorescent primers were performed on ABI sequencers, and ROX 500 (Applied Biosystems) was added as a standard size. Fragment analyses were performed using GENESCAN V (Applied Biosystems).

The genetic diversity analyses considered the 90 samples as a single bromeliad population as well as they divided by the host-tree species. We used Arlequin V3.1 program (Excoffier et al. 2006) to estimate the observed (HO = number of heterozygotes at a locus / total number of individuals surveyed) as the expected heterozygosity (HE = 1 - [SIGMA](pi)2, pi = allele frequency) and its standard errors following Nei's equations (1987). In addition, the allelic diversity was estimated (AD = total number of alleles found / total loci analyzed, Frankham et al. 2002). This software analyzes deviations from Hardy-Weinberg equilibrium following the procedure developed by Guo and Thompson (1992), based on the exact Fisher test. This program also computes the linkage equilibrium between each pair of loci based on a likelihood-ratio test (Slatkin and Excoffier 1996). In addition, a Bonferroni test correction was carried out, in order to test the significance of multiple comparisons of Hardy Weinberg and linkage equilibrium tests following the procedure described in Sokal and Rohlf (2003).

In order to describe the genetic structure, we estimated the gene flow between populations (Nm = 1-FST / 2FST, Wright 1951) and the genetic differentiation index (RST) adjusted for microsatellite data (Michalakis and Excoffier 1996). The inbreeding coefficient (FIS = HE-HO / HE, Nei 1977) was estimated for epiphytes from each host tree species. We compared the FIS index estimated in each population based on t test (Sokal and Rohlf 2003). In addition the total inbreeding coefficient for species T. recurvata was also calculated (FIT = FIS + [(FST) (1- FIS]), using the F indices modified from Wright (Nei 1977).

Results: Based on the analysis of 90 samples of Tillandsia recurvata, 12 alleles for the five microsatellite loci were found, and we estimated a mean HO of 0.42 [+ or -] 0.079 S.D and a mean HE of 0.82 [+ or -] 0.042. The mean allele number was relatively high (NA = 9.5 [+ or -] 0.70). The total inbreeding coefficient estimated for T. recurvata was relatively low (FIT = 0.25).

The bromeliad population collected from Prosopis laevigata showed lower genetic diversity than those from Parkinsoniapraecox (Table 1). In the bromeliads of P. laevigata the mean observed heterozygosity (HO = 0.34 [+ or -] 0.09 SD) was lower than the expected heterozygosity (HE = 0.79 [+ or -] 0.05), and the mean allelic diversity was of NA = 9.33 [+ or -] 0.94. According to the Bonferroni test, all loci were found in Hardy-Weinberg disequilibrium (P values < 0.05). In this bromeliad population a mean inbreeding coefficient showed a deficiency of heterozygotes (FIS = 0.57 [+ or -] 0.11 SD). In contrast, the bromeliad's population from P. praecox showed a mean observed heterozygosity (HO = 0.59 [+ or -] 0.06 SD) lower than the mean expected heterozygosity (HE = 0.86 [+ or -] 0.02). On the Bonferroni test all loci were in Hardy-Weinberg disequilibrium (P values < 0.05). The mean allelic diversity was of NA = 9.7 [+ or -] 0.5, and the mean inbreeding population coefficient (FIS = 0.31 [+ or -] 0.07), showed a deficiency of heterozygotes. The Bonferroni test found that all loci were in linkage equilibrium in the bromeliad populations collected in both P. laevigata and P. praecox.

The gene flow between populations established in the two host-tree species was high (Nm = 16) and consequently they are not genetically differentiated (Rst = 0.03, P = 0.09).

Discussion and conclusions: Although our results found a low genetic diversity of Tillandsia recurvata, these do not support the idea that it is an autogamous species, as it was suggested by Soltis et al. (1987). These authors based this idea from the zero genetic diversity found in the populations studied with allozymes markers. In contrast, the study of Ramirez-Padilla (2008) carried out with microsatellite markers found a relatively high observed genetic diversity (HO = 0.67), and an excess of heterozygotes (FIS = -0.028). The zero genetic diversity found by Soltis et al. (1987) was probably caused by the molecular technique used (allozymes), since it detects lower genetic variation than microsatellite markers (Agarwal et al. 2008). In addition, the sampling method used could affect the genetic diversity results found between Soltis et al (1987) and our study. We collected each plant from a different host tree (i.e. 90 bromeliads of 90 different host trees) in contrast Soltis et al. (1987) collected 243 plants, but only from seven different host plants. Hence, it is highly probable that the clumps collected by Soltis et al (1987) represented closest relatives because the vegetative reproduction is most frequent that the sexual reproduction in this epiphyte.

The three population genetic studies carried out with Tillandsia recurvata do not show a clear trend about the genetic diversity contained in this species so it is not possible to infer a possible pattern in population genetic diversity for this species.

In addition, the genetic knowledge available for the epiphytes is scarce, though this life form represents about 10% of the world vascular flora (Madison 1977). Among other vascular epiphytic species that have received attention are the orchids, which have been analyzed mainly with isoenzymes (e.g. Ackerman and Ward 1999, Avila and Oyama 2007) and recently with microsatellites (e.g. Jacquemyn et al. 2009, Rodrigues and Kumar 2009). Other vascular epiphyte species have been studied with some other molecular marker, as the Brazilian herbaceous Monstera adansonii var. klotzschiana, which used two AFLPs markers (Andrade et al. 2007). All these studies found low levels of genetic diversity, similarly to the results of our study.

We found an unexpected lower genetic diversity in the bromeliads on Prosopis laevigata than those on Parkinsonia praecox. However, the former host species has a higher population density than in P. praecox and according to genetic population theory, the highest genetic diversity must be associated with the largest population size (Hedrick 2005). Probably, this unexpected result could be explained by processes occurred after seed germination or vegetative reproduction and human management type by the local communities.

Our study found that there is gene flow between the bromeliad populations established between the two tree species and they have different genetic diversity when are analyzed according to the tree species used. These results suggest that future ecologic studies would consider a metapopulation focus to define the populations of this bromeliad, as we as to search selective process after germination of the seeds that could be occurring according to the tree species used as phorophyte.


This study was supported by UNAM (UNAM, SDEI-PTID-02), and DIP-FESI (grant given to SS for SNI researchers, FES Iztacala-UNAM). Thank V Garcia, H. Tapia, U. Guzman and A. Ibarra for fieldwork support. The Figure 1 was drafted by Dr. O. Tellez, UBIPRO, UNAM. The electrophoretic analyses were run on 3100 sequencer of FES-Iztacala, UNAM (A. Monsalvo) and ABI 310 of IB-UNAM (L. Marquez). The ordinary primers were quickly synthesized by IBT, UNAM (P. Gaytan and E. Lopez).

Literature Cited

Ackerman, J. D. and S. Ward (1999). "Genetic variation in a widespread, epiphytic orchid: Where is the evolutionary potential?" SYST BOT 24: 282-291.

Agarwal, J. D. (2008). "Advances in molecular marker techniques and their applications in plant sciences." Plant Cell Rep. 27: 617-631.

Andrade, I. M., S. J. Mayo, et al. (2007). "A preliminary study of genetic variation in populations of Monstera adansonii var. klotzschiana (Araceae) from North-East Brazil, estimated with AFLP molecular markers." ANN BOT 100: 1143-1154.

Avila-Diaz, I. and K. Oyama (2007). "Conservation genetics of an endemic and endangered epiphytic Laelia speciosa (Orchidaceae)." AM J BOT 94: 184-193.

Bawa, K. S. and M. Hadley (1991). Reproductive ecology of tropical forests plants. Informa healthcare. New York, Informa Healthcare.

Beebee, T. and G. Rowe (2008). Molecular Ecology. New York, Oxford University Press.

Benzing, D. H. (1990). Vascular epiphytes: general biology and related biota. Cambridge UK, Cambridge University Press.

Benzing, D. H. and A. Renfrow (1971). "The Biology of the epiphytic bromeliad Tillandsia circinata Schlecht. I. The nutrient status of populations in South Florida." AM J BOT 58: 867-873.

Bernal, R., T. Valverde, et al. (2005). "Habitat preference of the epiphyte Tillandsia recurvata (Bromeliaceae) in a semi-desert environment in Central Mexico." CAN J BOT 83: 12381247.

Boneh, L., P. Kuperus, et al. (2003). "Microsatellites in the bromeliads Tillandsia recurvata and Guzmama monostachya." MOL ECOL NOT 3: 302-303.

Callaway, R. M., K. O. Reinhart, et al. (2002). "Epiphyte host preferences and host traits: mechanisms for species-specific interactions." OECOLOGIA 132: 221-230.

Excoffier, L., G. Laval, et al. (2006). An integrated software package for population genetics data analysis, version 3.01. Berne, Computational and Molecular Population Genetics Lab. Institute of Zoology, University of Berne.

Frankham, R., J. D. Ballou, et al. (2002). Introduction to conservation genetics. Cambridge UK, Cambridge University Press.

Garcia-Suarez, M. D., V Rico-Gray, et al. (2003). "Distribution and abundante of Tillandsia spp (Bromeliaceae) in the Zapotitlan valley, Puebla, Mexico." PLANT ECOL 166: 207-215.

Gonzalez-Astorga, J., A. Cruz-Angon, et al. (2004). "Diversity and genetic structure of the Mexican endemic epiphyte Tillandsia achyrostachys E. Morr. ex Baker var. achyrostachys (Bromeliaceae)." ANN BOT 94: 545-51.

Guo, S. and E. Thompson (1992). "Performing the exact test of Hardy Weinberg proportion for multiple alleles." BIOMETRICS 48: 361-372. Hedrick, P. (2005). Genetics of Populations 3rd ed. Tempe, AZ, Jones and Bartlett Publishers

Jacquemyn, H., R. Brys, et al. (2009). "Effects of population size and forest management on genetic diversity and structure of the tuberous orchid Orchis mascula." CONSERV GENET 10: 161-168.

Jarne, P. and P. J. L. Lagoda (1996). "Microsatellites, from molecules to populations and back. Trends in Ecology and Evolution." TRENDS ECOL EVOL 11: 424-429.

Loveless, M. D. and J. L. Hamricj (1984). "Ecological determinants of genetic structure in plant populations." ANN REV SYS 15: 65-95.

Madison, M. (1977). "Vascular epiphytes: their systematic occurrence and salient features." Selbyana 2: 1-13.

Michalakis, Y and L. Excoffier (1996). "A generic estimation of population subdivision using distances between alleles with special reference for microsatellite loci." GENETICS 142: 1061-1064.

Montana, C., R. Dirzo, et al. (1997). "Structural Parasitism of an Epiphytic Bromeliad upon Cercidium praecox in an Intertropical Semiarid Ecosystem." BIOTROPICA 29: 517-521.

Nei, M. (1977). "F. statistics and analysis of gene diversity in subdivided populations." ANN HUM GENET 41: 225-233.

Nei, M. (1987). Molecular evolutionary genetics. New York, Columbia University Press.

Pinto, R., I. Barria, et al. (2006). "Geographical distribution of Tillandsia lomas in Atacama Desert, northern Chile." J ARID ENVIRON 65(543-552).

Ramirez-Morillo, I., M. F. Gonzalez, et al. (2008). "Reproductive biology of six species of Tillandsia L. (Bromeliaceae) in Mexico." J. Bromeliad Soc. 58(4): 149-155.

Ramirez-Padilla, C. (2008). Analisis de la dispersion de semillas en una metapoblacion de la epifita Tillandsia recurvata L. (Bromeliaceae) a traves del uso de microsatelites. Mexico, Universidad Nacional Autonoma de Mexico. M. Sc.

Rodrigues, K. and S. V. Kumar (2009). "Isolation and characterization of 24 microsatellite loci in Paphiopedilum rothschildianum, an endangered slipper orchid." CONSERV GENET 10: 127-130.

Rzedowski, J. (1978). Vegetacion de Mexico. Mexico, Editiorial Limusa.

Slatkin, M. and L. Excoffier (1996). "Testing for linkage disequilibrium in genotypic data using the EM algorithm." HEREDITY 76: 377-383.

Sokal, R. and J. Rohlf (1996). Biometry. 3rd Ed. New York, Freeman Press.

Soltis, D. E., A. J. Gilmartm, et al. (1987). "Genetic variation in the epiphytes Tillandsia ionantha and Tillandsia recurvata (Bromeliaceae)." AM J BOT 74: 531-537.

Valiente-Banuet, V., A. Casas, et al. (2000). "La vegetacion del Valle de Tehuacan-Cuicatlan." BOL SOC BOT MEX.

Corresponding author:

Sofia Solorzano. UBIPRO, FES-Iztacala,UNAM. Avenida de los Barrios No. 1. Los Reyes Iztacala, Tlalnepantla, Estado de Mexico. 54090. Mexico. Phone: +52 (55) 56 23 11 29, Fax: +52 (55) 56 23 12 25. E-mail:

Sofia Solorzano, Sandra J. Solfs & Patricia Davila. (1)

(1) UBIPRO, FES-Iztacala,UNAM. Avenida de los Barrios No. 1. Los Reyes Iztacala, Tlalnepantla, Estado de Mexico.54090. Mexico.
Table 1. Locus name, sequence and flurochrome the primers used to
amplify microsatellite regions in Tillandsia recurvata. The Ta is
the alignment temperature adjusted to this study species. The
column NA represents the mean allele number by locus for the species.
The mean observed (HO) and expected (HE) gene diversity are showed
according to the host tree species were the bromeliads were collected.

name    Primer sequence (5'-3')         Ta ([degrees]C)   NA

CT5     F-HEX:                          51                10

e6      F-FAM: AAACTATGGATTCCCCAACT     54                9

e19     F-FAM TCTTACTGCTCTCCATTGGT      52                3

e6b     F-HEX: CGTACGAAGGTAAGCACAA      51                2

p2p19   F-HEX: ATGCTGCCCACTGAAGATTT     51                11

        aP. laevigata           P. praecox
name    [H.sub.o]   [H.sub.e]   [H.sub.o]   [H.sub.e]

CT5     0.47        0.86        0.67        0.88

e6      0.30        0.76        0.53        0.84

e19     0.0004      0.03        0.00        0.35

e6b     0.00        0.45        0.00        0.28

p2p19   0.25        0.75        0.57        0.85
COPYRIGHT 2010 Bromeliad Society International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Scientific
Author:Solorzano, Sofia; Solis, Sandra J.; Davila, Patricia
Publication:Journal of the Bromeliad Society
Geographic Code:100NA
Date:Mar 1, 2010
Previous Article:Studies on Orthophytum--Part XI: three new species from Bahia and Minas Gerais.
Next Article:Dyckia estevesii revisited.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters