全 文 ：Journal of Systematics and Evolution 46 (4): 595–599 (2008) doi: 10.3724/SP.J.1002.2008.07048
(formerly Acta Phytotaxonomica Sinica) http://www.plantsystematics.com
Systematic position of Gomphogyne (Cucurbitaceae) inferred from ITS,
rpl16 and trnS-trnR DNA sequences
1,2Hong-Tao LI 1De-Zhu LI ∗
1(Key Laboratory of Biodiversity and Biogeography, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming 650204, China)
2(The Graduate School of Chinese Academy of Sciences, Beijing 100049, China)
Abstract This paper assessed the systematic position of the genus Gomphogyne. The nuclear ITS, the chloro-
plast rpl16, and trnS-trnR sequences were used to reconstruct the phylogeny of Gomphogyne and its related gen-
era. Analyses of three separate and combined datasets provided a good amount of informative characters and re-
solved the systematical relationships of Gomphogyne well. The maximum parsimony analyses revealed that: (1)
Gomphogyne was a natural genus and was different from the genera Hemsleya and Gynostemma; (2) Hemsleya
delavayi and H. macrocarpa did not belong to the genus Gomphogyne, but to the genus Hemsleya; and (3)
Gomphogyne was sister to Hemsleya. It was assured that Gomphogyne was a monotypic genus. These results were
largely in agreement with the systems of classification of the Cucurbitaceae of Jeffrey in 1990 and in 2005 and
with that of Li for Hemsleya in 1993, but were different from the previous studies in which Gomphogyne and Gy-
nostemma together were suggested to be sister to Hemsleya.
Key words Gomphogyne, ITS, molecular phylogeny, rpl16, trnS-trnR.
Gomphogyne Griff., a member of subtribe
Gomphogyninae of tribe Zanonieae (Cucurbitaceae)
(Jeffrey, 1990, 2005), was established in 1841 by
Griffith with one species G. cissiformis Griff. Hence-
forth, a few new species were reported, namely, G.
delavayi Gagnep., G. macrocarpa Cogn. and G. het-
erosperma (Wall.) Kurz. However, later studies re-
vealed that all the new species belonged to the genus
Hemsleya Cogn., in which they were named as H. de-
lavayi (Gagnep.) C. Jeffrey ex C. Y. Wu & C. L.
Chen, H. macrocarpa (Cogn.) C. Y. Wu ex C. Jeffrey
and H. heterosperma (Wall.) C. Jeffrey. As a result,
the genus Gomphogyne became a monotypic genus
comprising only one species, i.e., the type species, G.
cissiformis (Li, 1993; Lu et al., 2007). It is principally
an Indo-Malayan genus with its northernmost distri-
bution in China. Besides Gomphogyne, subtribe
Gomphogyninae also harbored Hemsleya and Gy-
nostemma and they formed a monophyletic group
(Jeffrey, 1990, 2005; Li, 1993). Neoalsomitra Hutch.
or Neoalsomitra and Zanonia L. together were sug-
gested to be the sister to the subtribe (Li, 1993). Hem-
sleya (Li, 1993) comprised about 24 species and Gy-
nostemma (Chen, 1995; He, 1996) consisted of 17
species. The genus Gomphogyne is characterized by
monoecia and differentiates the dioecism of Hemsleya
and Gynostemma. The male flowers possess five
stamens, while female flowers bear three stigmas bi-
furcating at the apex, in a raceme or panicle. Five
densely imbricate petals build the rotatable pale green
corollas. Fruits are capsular, turbinate, venose and
ribbed. The plants are herbaceous vines, with pal-
mately compound and alternate leaves comprising
seven leaflets (Lu, 1986).
As Gomphogyne is monotypic, there are few
studies involving the genus. We can only acquire data
of Gomphogyne from the studies on Cucurbitaceae.
Moreover, all the data focused mainly on either mor-
phology (Lu, 1986) or chromosome numbers (Singh,
1990). Two questions remained unanswered. First,
whether the taxon should be recognized as an inde-
pendent genus; second, whether it is sister to Hem-
sleya as suggested earlier (Li, 1993) or not. Although
studies of molecular phylogeny of many genera of the
Cucurbitaceae have been comprehensively carried out,
Gomphogyne has not yet been included in previous
studies (Sanjur et al., 2002; Clark et al., 2006).
This study addressed to the systematic position of
Gomphogyne. We sequenced the nuclear ITS regions,
the chloroplast rpl16 intron and flanking regions, and
the chloroplast trnS-trnR regions of the genus and its
relatives. These sequences have been useful in clari-
fying phylogenetic relationships in the Cucurbitaceae
(Jobst et al., 1998; Jarret & Newman, 2000; Gar-
cia-Mas et al., 2004). The major objective of this pa-
per is to confirm the systematic position of
Gomphogyne, particularly its relationships with Hem-
sleya and Gynostemma.

———————————
Received: 19 March 2007 Accepted: 8 August 2007
* Author for correspondence. E-mail: .
Journal of Systematics and Evolution Vol. 46 No. 4 2008 596
1 Material and methods
1.1 Plant samples
For this study, the only Gomphogyne species and
representatives of the remaining genera in subtribe
Gomphogyninae (Jeffrey, 1990, 2005; Li, 1993; Chen,
1995; Kocyan et al., 2007), Hemsleya and Gy-
nostemma, were sampled including the type species of
Gynostemma. As Hemsleya is a monophyletic group
based on morphological evidence (Li, 1993) and mo-
lecular data (Li, 2007), five Hemsleya species were
sampled, including the type species, and H. delavayi
and H. macrocarpa that were previously regarded as
Gomphogyne species. Hemsleya heterosperma was
not sampled because of inaccessibility. We used a
species of Neoalsomitra, which was suggested to be
sister to subtribe Gomphogyninae (Li, 1993), as out-
group to root the tree. Samples were all obtained from
wild plants (Table 1). Healthy, clean leaves were col-
lected and quickly dried in silica gel, with voucher
herbarium specimens deposited at the Herbarium of
the Kunming Institute of Botany, Chinese Academy of
Sciences (KUN).
1.2 DNA extraction, amplification and sequencing
For each sample, total genomic DNA was iso-
lated from 0.2 g silica-gel-dried or 0.4 g fresh leaves
using the modified CTAB method (Doyle & Doyle,
1987), with 4% CTAB instead of 2% CTAB. Leaf
tissue was ground in liquid nitrogen before using
CTAB.
Following extraction, DNA was amplified using
the polymerase chain reaction (Saiki et al., 1988). The
ITS region includes the ITS1, 5.8S, and ITS2 nuclear
rDNA regions. It was amplified with primers ITS-4
(5′-TCC TCC GCT TAT TGA TAT GC-3′) and ITS-5
(5′-GGA AGG AGA AGT CGT AAC AAG G-3′)
(White et al., 1990). The rpl16 intron region was am-
plified as described in Jordan et al. (1996) with prim-
ers rpl16-F71 (5′-GCT ATG CTT AGT GTG TGA
CTC GTT G-3′) and rpl16-R1516 (5′-CCC TTC ATT
CTT CCT CTA TGT TG-3′). The trnS-trnR region
was amplified with primers trnSGCU-F (5′-CGC CGC
TTT AGT CCA CTC A-3′) (Doyle et al., 1992) and
trnR-R (5′-ATT GCG TCC AAT AGG ATT TGA
A-3′) (Dumolin-Lapegue et al., 1997). The thermal
cycler (PE9600 or PE9700) for trnS-trnR was pro-
grammed for an initial step of 4 min at 94 , fo℃ l-
lowed by 36 cycles of 1 min at 94 , 90℃ s at 50 , 90℃
s at 72 , and a final extension of 7 min at 72℃ . The ℃
PCR products were visualized by agarose gel electro-
phoresis, cleaned with Wizard PCR preps DNA Puri-
fication system (Promega, Madison, WI, USA), and
both strands were sequenced using the same primer
combination as for PCR amplifications.
1.3 Phylogenetic analyses
Clustal X (Thompson et al., 1997) was used to
produce an aligned matrix, which was corrected
manually using the BioEdit program (Hall, 1999). The
indels were coded using GapCoder method (Simmons
& Ochoterena, 2000) in the data matrix.
Maximum parsimony (MP) analyses were con-
ducted using PAUP 4.0b10 (Swofford, 2001). Char-
acters were treated as unordered and unweighted.
Heuristic searches were conducted with
Tree-Bisection Reconnection (TBR) branch swapping,
MulTrees ON, and 1000 random taxon addition repli-
cates holding 20 trees at each step. Branch support
(BS) values for individual clades were calculated by
running 1000 bootstrap replicates of the data, with
starting trees acquired by a single replicate of random
stepwise addition of taxa, under TBR branch swap-
ping, and MulTrees ON. The consistency index (CI),
retention index (RI) and rescaled consistency index
(RC) were obtained with PAUP 4.0b10.
Incongruence among multiple data partitions
(ITS, rpl16 and trnS-trnR) was evaluated with the par-
tition homogeneity test (Farris et al., 1994, 1995) im-
plemented in PAUP 4.0b10 (Swofford, 2001). The
partition homogeneity test used 1000 resampling rep-
licates under the maximum parsimony criterion, and
all characters were equally weighted.
2 Results
2.1 Sequence comparisons
Sequences information for the four datasets (ITS,
rpl16, trnS-trnR and the combined ITS+rpl16+trnS-
trnR) of this study is shown in Table 2. The aligned
ITS region has a length of 703 characters, with 139
(19.8%) variable sites, 65 (9.2%) of which were par-
simony-informative. The aligned rpl16 sequence has
1062 characters, with 97 (9.1%) variable characters,
34 (3.2%) were parsimony-informative characters.
The aligned trnS-trnR sequence has 1701 characters,
132 (7.8%) of which were variable characters, and 39
(2.3%) were parsimony-informative characters.
2.2 Phylogenetic analyses
Separate analyses of the ITS, rpl16 and trnS-trnR
datasets produced well resolved trees. The partition
homogeneity test suggested that the ITS, rpl16 and
trnS-trnR datasets were not significantly incongruent
(P=1). At the same time, each of three dataset pro-
duced identical topologies, thus these three datasets
were combined for phylogenetic analyses. The aligned
combined sequences including coded gaps had 3466
LI & LI: Systematic position of Gomphogyne inferred from molecular data

597
Table 1 The list of taxa and voucher specimens used in this study.
GenBank # Locality information Voucher Species
ITS rpl16 trnS-trnR
Gomphogyne cissiformis Griff. (锥形果) EF621663 EF621641 EF621684 Yongde, Yunnan, China (云南永德) H. T. Li (李洪涛) 835 (KUN)
Gynostermma pentaphyllum (Thunb.)
Makino. (绞股蓝) EF621662 EF621622 EF621683 Huangshan, Anhui, China (安徽黄山) H. T. Li (李洪涛) 005 (KUN)
Hemsleya graciliflora (Harms) Cogn. (马
铜铃) EF621654 EF621640 EF621675 Pengzhou, Sichuan, China (四川彭州) H. T. Li (李洪涛) 069 (KUN)
Hemsleya delavayi (Gagnep.) C. Jeffrey
ex C. Y. Wu & C. L. Chen. (短柄雪
胆)
EF424063 EF424072 EF424080 Songming, Yunnan, China (云南嵩明) H. T. Li (李洪涛) 048 (KUN)
Hemsleya macrocarpa (Cogn.) C. Y. Wu
ex C. Jeffrey (圆锥果雪胆) EF621652 EF621632 EF621672 Yongde, Yunnan, China (云南永德) H. T. Li (李洪涛) 003 (KUN)
Hemsleya lijiangensis Lu ex C. Y. Wu &
C. L. Chen. (丽江雪胆) EF424065 EF424075 EF424078 Lijiang, Yunnan, China (云南丽江) H. T. Li (李洪涛) 047 (KUN)
Hemsleya chinensis Cogn. (雪胆)

EF424064

EF424073

EF424081

Mt. Emei, Sichuan, China
(四川峨眉山)
H. T. Li (李洪涛) 023 (KUN)

Neoalsomitra integrifoliola (Cogn.)
Hutch. (棒锤瓜)
EF621642

EF621620

EF621664

Xishuangbanna, Yunnan, China
(云南西双版纳)
H. T. Li (李洪涛) 803 (KUN)


Table 2 DNA site variation and tree statistic for the four data sets used in the phylogenetic analyses of taxa presented in this study
Number of char-acters
Number of vari-
able sites
Number of in-
formative sites
No. trees Tree length CI RI RC
ITS 703 139 65 1 275 0.916 0.744 0.682
rpl16 1062 97 34 1 144 0.972 0.915 0.889
trnS-trnR 1701 132 39 1 178 0.962 0.865 0.833
ITS + rpl16+ trnS-trnR 3466 368 138 1 606 0.941 0.810 0.761
CI, consistency index; RC, rescaled consistency index; RI, retention index.

characters, with 368 (10.6%) variable and 138 (4%)
parsimony-informative sites.
In our study, the MP analyses of four datasets
each produced only a tree and the resulting topologies
were identical (Figs. 1–4). All MP analyses recovered
well-resolved and strongly supported topologies. The
MP analysis of combined data yielded one most par-
simonious tree of 606 steps (CI=0.941, RI=0.810,
RC=0.761). The tree was well resolved with strong
branch support (Fig. 1), as shown by the bootstrap
values. In the analyses three clades were identified
(Fig. 1), all with strong bootstrap support (BS=
100%). Each clade represents a different genus of
subtribe Gomphogyninae. The basal group was G.
pentaphyllum representing the genus Gynostemma.
The monotypic genus Gomphogyne followed as sister
to the genus Hemsleya.
3 Discussion
ITS, rpl16 and trnS-trnR sequences were useful
for resolving the intergeneric even interspecific rela-
tionships in this study. The separate and combined
analyses of the three DNA regions all recovered
well-resolved and strongly supported trees, in which
the combined tree was most strongly supported.
Our study revealed that H. delavayi and H. mac-
rocarpa were members of Hemsleya, though they
were previously treated as Gomphogyne species. It
corresponded to Li’s (1993) classification of Hem-
sleya, and consequently, it was assured that
Gomphogyne was a monotypic genus. In comparison
with rest of the taxa sampled in the study,
Gomphogyne had many indel singletons in the three
DNA regions. There were five indels ranging from 1
to 5 bps in the ITS sequences, and five indels of 1 bp
in the rpl16 intron, and four indels in trnS-trnR, one of
which was a 99-bp long deletion. These indel single-
tons are the unique molecular characteristics to
Gomphogyne.
In the separate and combined DNA analyses, the
MP systematic trees presented three main strongly
supported clades (Fig. 1). Moreover, each clade rep-
resented a different genus of the subtribe Gom-
phogyninae, respectively. These relationships also
corresponded to the previous classification (Jeffrey,
1990, 2005). Certainly, the results of our analyses
suggested a different scheme of relationships. Only
the Gomphogyne was sister to Hemsleya. It did not
corroborate the results of previous studies, such as Li
(1993), in which Gomphogyne and Gynostemma to-
gether were suggested to be sister to Hemsleya.
Journal of Systematics and Evolution Vol. 46 No. 4 2008 598






























Figs. 1–4. Single most parsimonious trees of Gomphogyne and related genera based on combined data of three DNA regions (Fig. 1), ITS (Fig. 2),
rpl16 (Fig. 3) and trnS-trnR separate data sets (Fig. 4). Numbers above the lines are bootstrap values. Numbers below the lines are branch lengths.

The phylogeny of subtribe Gomphogyninae pre-
sented here for the first time provided a valuable pic-
ture of the relationships among the three genera, i.e.,
Gomphogyne, Hemsleya and Gynostemma, of the sub-
tribe Gomphogyninae as defined by Jeffrey. Further
studies with increased species sampling, especially of
more Gynostemma species, are needed to ultimately
resolve the relationships of Gomphogyne among the
three genera of the subtribe Gomphogyninae.
Acknowledgements This work was supported by
grants from the Ministry of Science and Technology
(2005DKA21006) to De-Zhu LI, and the Yunnan Pro-
vincial Natural Science Foundation (2005C00- 50M)
to Lian-Ming GAO.
References
Chen S-K ( 陈书坤 ). 1995. A classificatory system and
geographical distribution of the genus Gynostemma Bl.
(Cucurbitaceae). Acta Phytotaxonomica Sinica (植物分类
学报) 33: 403–4l0.
Clarke AC, Burtenshaw MK, McLenachan PA, Erickson DL,
Penny D. 2006. Reconstructing the origins and dispersal of
the Polynesian bottle gourd (Lagenaria siceraria).
Molecular Biology and Evolution 23: 893–900.
Doyle JJ, Davis JI, Soreng RJ, Garvin D, Anderson MJ. 1992.
Chloroplast DNA inversions and the origin of the glass
family (Poaceae). Proceedings of the National Academy of
Sciences USA 89: 7722–7726.
Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for
small quantities of fresh leaf tissue. Phytochemical
Bulletin 19: 11–15.
Dumolin-Lapegue S, Pemonge MH, Petit RJ. 1997. An
enlarged set of consensus primers for the study of
organelle DNA in plants. Molecular Ecology 6: 393–398.
Farris JS, Källersjö M, Kluge AG, Bult C. 1994. Testing
significance of incongruence. Cladistics 10: 315–319.
Farris JS, Källersjö M, Kluge AG, Bult C. 1995. Constructing a
significance test for incongruence. Systematic Biology 44:
570–572.
Garcia-Mas J, Monforte AJ, Arús P. 2004. Phylogenetic
relationships among Cucumis species based on the
LI & LI: Systematic position of Gomphogyne inferred from molecular data

599
ribosomal internal transcribed spacer sequence and
microsatellite markers. Plant Systematics and Evolution
248: 191–203.
Hall TA. 1999. BioEdit: a user-friendly biological sequence
alignment editor and analysis program for Windows
95/98/NT. Nucleic Acids Symposium Series 41: 95–98.
He S-Z (何顺志). 1996. A new species of Gynostemma Bl.
(Cucurbitaceae) from Guizhou, China. Acta Phytotaxono-
mica Sinica (植物分类学报) 34: 207–209.
Jarret RL, Newman M. 2000. Phylogenetic relationships among
species of Citrullus and the placement of C. rehmii De
Winter as determined by internal transcribed spacer (ITS)
sequence heterogeneity. Genetic Resources and Crop
Evolution 47: 215–222.
Jeffrey C. 1990. Systematics of the Cucurbitaceae: An
overview. In: Bates DM, Robinson RW, Jeffrey C eds.
Biology and utilization of the Cucurbitaceae. Ithaca &
London: Comstock Publishing Associates, Cornell
University Press. 3–28.
Jeffrey C. 2005. A new system of Cucurbitaceae. Botanicheskij
Zhurnal 90: 332–335.
Jobst J, King K, Hemleben V. 1998. Molecular evolution of the
internal transcribed spacers (ITS1 and ITS2) and
phylogenetic relationships among species of the family
Cucurbitaceae. Molecular Phylogenetics and Evolution 9:
204–219.
Jordan WC, Courtney MW, Neigel JE. 1996. Low levels of
intraspecific genetic variation at a rapidly evolving
chloroplast DNA locus in North American duckweeds
(Lemnaceae). American Journal of Botany 83: 430–439.
Kocyan A, Zhang L-B, Schaefer H, Renner SS. 2007. A
multi-locus chloroplast phylogeny for the Cucurbitaceae
and its implications for character evolution and
classification. Molecular Phylogenetics and Evolution 44:
553–577.
Li D-Z (李德铢). 1993. Systematics and evolution of Hemsleya
(Cucurbitaceae). Kunming: Yunnan Science & Technology
Press.
Li H-T (李洪涛). 2007. Molecular phylogeny and biogeo-
graphy of Hemsleya (Cucurbitaceae). Ph.D. Dissertation.
Kunming: Kunming Institute of Botany, Chinese Academy
of Sciences.
Lu A-M ( 路安民 ). 1986. Gomphogyne Griff. In: Flora
Reipublicae Popularis Sinicae (中国植物志 ). Beijing:
Science Press. 73 (1): 96–98.
Lu A-M (路安民), Chen S-K (陈书坤). 1986. Gynostemma Bl.
In: Flora Reipublicae Popularis Sinicae (中国植物志).
Beijing: Science Press. 73 (1): 265–277.
Lu A-M (路安民), Jeffrey C. 2007. Flora of China (online).
Available from http://www.fna.org/china/mss/volume19/C
ucurbitaceae-MO_coauthoringFeb2007.htm.
Saiki RK, Gelfand DH, Stoffel S, Scharf SJ, Higuchi R, Horn
GT, Mullis KB, Erlich HA. 1988. Primer-directed
enzymatic amplification of DNA with a thermostable
DNA-polymerase. Science 239: 487–491.
Sanjur OJ, Piperno DR, Andres TC, Wessel-Beaver L. 2002.
Phylogenetic relationships among domesticated and wild
species of Cucurbita (Cucurbitaceae) inferred from a
mitochondrial gene: Implications for crop plant evolution
and areas of origin. Proceedings of the National Academy
of Sciences USA 99: 535–540.
Simmons MP, Ochoterena H. 2000. Gaps as characters in
sequence based phylogenetic analyses. Systematic Biology
49: 369–381.
Singh AK. 1990. Cytogenetics and evolution in the
Cucurbitaceae. In: Bates DM, Robinson RW, Jeffrey C
eds. Biology and utilization of the Cucurbitaceae. Ithaca &
London: Comstock Publishing Associates, Cornell
University Press.
Swofford DL. 2001. PAUP*. Phylogenetic analysis using
parsimony (*and other methods). Version 4. Sunderland,
MA: Sinauer Associates.
Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins
DG. 1997. The CLUSTAL_X windows interface: flexible
strategies for multiple sequence alignment aided by quality
analysis tools. Nucleic Acids Research 25: 4876–4882.
White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification and
direct sequencing of fungal ribosomal RNA genes for
phylogenetics. In: Innis M, Gelfand D, Sninsky J, White T
eds. PCR protocols: A guide to methods and application.
San Diego: Academic Press. 315–322.
基于 ITS、rpl16 和 trnS-trnR DNA 序列讨论
锥形果属的系统位置
1,2李洪涛 1李德铢∗
1(中国科学院昆明植物研究所生物多样性与生物地理学重点实验室 昆明 650204)
2(中国科学院研究生院 北京 100049)

摘要 基于核基因ITS和叶绿体基因rpl16、trnS-trnR的DNA序列讨论了锥形果属的系统位置, 3个基因片段独立以及联合的分
析为锥形果属Gomphogyne的系统进化研究提供了足够的信息。结果表明: (1)锥形果属是一个自然属; (2)雪胆属Hemsleya的短
柄雪胆H. delavayi和圆锥果雪胆H. macrocarpa曾经被作为锥形果属的种, 分子证据表明它们确实隶属于雪胆属; (3)锥形果属
单独构成雪胆属的姊妹群, 而并非是与绞股蓝属Gynostemma共同构成。
关键词 锥形果属; ITS; 分子系统学; rpl16; trnS-trnR