a #j a Acla Genetica Sinku, July 2006, 33 (7): 59G597 ISSN 0379-4172

Molecular Evolution of Prolactin Gene Family in Rodents

LI ~ i n g ' , * * ~ , ZHANG Ya-Ping1v2'0

1. Laboratory of Cellular and Molecular Evolution, and Molecular Biology of Domestic Animals, Kunming Institute of Zoology,

Chinese Academy of Sciences, Kunming 650223, China;

2. Laboratory for Conservation and Utilization of Bio-resources, Yunnan University, Kunming 650091, China;

3. The Graduate School, Chinese Academy of Sciences, Beijing 100039, China

Abstract: In this study, we identified two novel members of prolactin gene family in rat by blast searches against the published genomic database. A further analysis showed that gene duplications leading to PRL gene family in rodents occurred after rodents diverged from other mammals. Major reorganization of the gene loci in rodents was largely completed before the split of rat and mouse. But PL-I and PL-I1 genes are the exceptions, which have clustered in a species-specific manner in the phylogenetic tree. By combining results from gene conversion testing, relative chromosomal location comparison and estimated time for gene duplication, we believe that rodent PL-I and PL-I1 genes are species-specific and are the results of serial duplications which occurred after the divergence of mouse and rat. Our analysis also reveals that continual gene duplication and divergence occurred during the evolution of rodent PRL gene family. Key words: prolactin; rodent; independent gene duplication Abbreviations: PLF, proliferin; PLF-RP, proliferin-related protein; PL-I or 11, placental I or 11; PLP, prolactin-like protein; PRP, prolactin-related protein

Prolactin along with growth hormone (GH), placental lactogen and related proteins in mammals form a hormone gene superfamily, which possesses a similar gene structure and function ", *I. In primates, the placental lactogens are considered to originate by tandem gene duplication of the GH gene ['I. However, in rodents, the placenta lactogens are more similar to pituitary prolactin than to growth hormone '31.

Prolactin gene family consists of multiple mem- bers in rodentia. These genes are expressed in a cellu- lar- and temporal-specific manner in the anterior pitui- tary, placenta, uterus [41, and are mainly involved in the process of reproduction and pregnan~y'~'. Based on the biological activities of the members, PRL gene family can be divided into two groups: classical members (in- cluding PRL, PL-I and PL-II genes), which use the PRL-receptor signaling pathway, and nonclassical

members, which use other pathways [41. Among these identified members, different exodintron organizations have been described. The prototypical members con- tain 5 exons and 4 introns, but a group of genes ac- quired one or two extra exons between exon 2 and exon 3 of the prototypical mernber~'~'. In mouse and rat, PRL gene family members are clustered on chro- mosomes 13 and 17, re~pectively'~-~'. A recent research using blast searches against the published mouse ge- nome database shows that at least 26 prolactin gene family members exist in mouse[61, but no similar analysis in rat has been reported.

In the present study, we used the published rat genomic information as a tool and identified two new members of the rat PRL gene family. Adding other available members of the rodent PRL gene family, we reconstructed the phylogenetic tree of this family. A

Received: 2005-10-1 8; Accepted: 2005- 11 -2 1

This work was supported by National Natural Science Foundation of China (No.30021004, 30430110) and the "Light in Western China" of the Chinese Academy of Sciences to Ying Li.

@ Corresponding author. E-mail: zhangyp@public.km.yn.cn; Tel: +86-871-519 0761, Fax: +86-871-519 5430

LI Ying et al.: Molecular Evolution of Prolactin Gene Family in Rodents 59 1

further analysis shows that independent gene duplica- tions occurred in PL-I and PL-I1 subfamilies after divergence of deer mouse, mouse and rat.

1 Materials and Methods

1.1 Sequence acquisition

New members of the prolactin gene family in rat were obtained by blast searches against the rat ge- nomic database published in June 2003 (http://genome.cse.ucsc.edu/cgi-bin/hgBlat) using all known amino acid and cDNA sequences of PRL gene family in mouse and rat. Genomic sequences con- taining similar sequences with members of rodent PRL family were extracted and aligned with the mRNA by a tool available at the website of NCBI (http://www.ncbi.nlm.nih.gov/IEB/Research/OstelVS pidey/spideyweb.cgi). This allows inspection of the exodintron pattern of each gene and extraction of coding sequences of the new members.

1.2 Evolution analyses

Sequences were aligned using Clustal X pro- gram [*] followed by manual adjustments. DNA se- quences retrieved from GenBank were also used in the analysis (Table 1). Maximum-parsimony (MP) tree was constructed by using amino acid distance implemented in the program PAUP 19’. Gaps were treated as the 21st states, and 500 replications of bootstrap analyses were done to the consensus tree to evaluate reliability. Geneconv 1.81 was used to test possible gene conversion events [lo’ with default set- tings. To test possible inconsistent rate of synony- mous substitution of those classic members, the method of Li and Bousquet [‘‘I implemented in RRTree were used for relative rate test. Linearized tree was constructed in Mega 2.1 [I3’.

2 Results

2.1 Blast analysis of rat genome

All cDNA sequences of mouse and rat PRL-related genes available in GenBank were used to search the rat genomic database. We identified totally

21 PRL-related genes, all of which were located in chromosome 17 with the entire locus spanning about 1.5 megabases. All identified rat PRL members and their chromosomal locations are shown in Table 1 and their exodintron organizations are shown in Fig. 1.

Among these loci, only two are novel. One se- quence shows identical exodintron structure with rat PL-I and share 99.5% cDNA sequence identity to rPL-I and 91.7% to rPL-IV. We thus named it rPL-IV2. The aligned amino acid sequences of these three genes are shown in Fig. 2. Another sequence shows 86.6% nucleotide similarity with the mPLP-E gene, both of which share the same unique 7 exod6 intron organiza- tion pattern. But we found a single nucleotide deletion and a ‘GT’ to ‘AT’ mutation at the 5’ splice site of exon4-intron4 boundary. These mutations may lead to shift of the open reading kame and alternative splicing, and thus this sequence is probably a pseudogene. Hence, we named it rPLP-EY. The sequence is not used in our analyses below but it is available upon request.

We note that loci with greater sequence and exodintron organization similarities tend to cluster in closer proximity in chromosomes as is the case in mouse ‘61, suggesting tandem duplication as a major expansion mechanism for this gene family. We also ob- served that almost all orthologous genes exist in both the rat and the mouse. This suggests that most of the reor- ganization of the loci in rodents has been completed before the divergence of rat and mouse. But rPL-N gene is an exception. It is not located closer to other PL-I genes but rather to the rat PLP-C genes. The cause of this unusual distribution pattern is not clear.

In addition, similar searches have also been done to the published human, chimpanzee and dog ge- nomes. The result shows only one PRL gene identi- fied for each species. And it has been suggested that glires is a sister taxon to primate [14]. This indicates that a cluster of PRL-related genes is rodent-specific and the expansion occurred after rodent diverged from other mammalian orders.

2.2 Phylogenetic tree

Rodent PRL famiIy members diverged so fast that the differences between those paralogous genes

592 B#%% Acta Genetica Sinica Vo1.33 No.7 2006

Table 1 Sequences used in this study

Gene Accession No. Gene start End Accession No.

mPRL

mPL-I a

mPL-I p mPL-I y

mPLP-J

mPL- II mPLP-I

mPLP-B

d P R P

mPLP-K

mPLP-Ca

mPLP-Cy

mPLP-CP

mPLP-CG

mPLP-N

mPLP-E

mPLP- F

mPLF-RP

mPLP-0

mPLP-M

mPLFl

mPLF2

mPLF3

mPLF4

mPLP-A

mPLP-L

dmPL-I

dmPL-IV

X02892

M35662, AF525 162

AF525 160

AF525 16 1

ABOl9 1 18

M14647

AF525 154

AF015563

AF015729

AF525 155

AF090140

AF466 150

AFl58744

AF525 158

AF525 156

AF0205 25

AF020524

X02594

AF525 157

AF234636

KO2245

KO3235

Nk-011954

AF128884

AFOl5562

AF2266 1 1

Vrana er a[. 2001[~~1

Vrana et a/. 2001"~l

rPRL

rPL-N2

rPL-I

rPLP-J

rPL- I1 rPLP-I

rPLP-B

rDPRP

rPLP-K

rPLP-D

rPLP-C

rPLP-H

rPL-IV

rPLP-CP

rPLP-N

rPLP-E

rPLP- F

rPLF-RP

rPLP-M

rPLP-A

rPLP-L

ghPL- rr ghPRL

mouse GH

rat GH

dmPL- IIA

dmPL- IIB

44711811

44569080

44435976

44386712

44345452

44289037

44275044

442 15670

44183876

4415 1357

440558 13

44019154

4391 8487

43880990

43786221

43686606

4366163 1

43622339

43419685

43286978

43 129367

I

I

I

I

I

I

dmPL- IIC 1

44702289

44563539

44430472

4438 1154

44340181

44307036

44270877

442059 16

44190749

44156911

4406 1479

44025045

439 12021

43886471

4379429 1

43764201

43674 198

43628650

43474057

4327939 1

43122303

I

I

I

I

I

I

I

AF022935

This study

D21103, M55269

AF150741, AB019945

M13749

ABOl9791

M31155

LO6441

AB022882, AF234635

AB000107

NM-173110

AB009889

U32679

AF239745

AF525 159

This study

AF139808

AFl39809, AF226609

A F 2 2 6 6 0 8

M13750

AF022883, AF226607

M27 146

S66296

NM008 11 7

V01237

Vrana et UE. 2001"~'

Vrana er al. 2001"~l

Vrana er al. 2001"~~

m: mouse; r: rat; drn: deer mouse; gh: golden hamster.

are much greater than those between most of the mammalian pituitary PRLs. So we used mouse and rat GH genes, which show about 25% amino acid sequence identity with the pituitary prolactin gene 'I5', as outgroups in our phylogenetic analysis.

All available rodent amino acid sequences were aligned using Clustal X [8'. We also added an ancestor mammalian PRL sequence (downloaded from Wallis homepage: http ://ww w. biols . susx . ac . uk/home/Mike- Wallis/Prolactin/MatPRL.html) when doing phy-

logenetic analysis. From our gene tree (Fig. 3), we observed that most of the genes possess orthologous members in mouse and rat, including rPLP-D VS

mPLP-a, PLP-Cy gene, dPRP gene, PLP-N, PLF-RP,

PLP-E, PLP-F; PLP-L, PLP-B, PRL, PLP-A, PLP-K, PLP-M, PLP-I, PLP-J and PL-I genes. A small pro- portion of the genes were identified only in mouse, with the rat counterparts becoming either absent or pseudogenes, such as PLP-0 , PLF and PLP-E genes. Most interestingly, when PL-I subfamily was con-

LI Ying et al.: Molecular Evolution of Prolactin Gene Family in Rodents 593

Chr. 17 1 44711811 r p ~ I

~ P L - I V ~

rPL-I

rPLP-J I rPL-II rPLP-I I

rPLP-B I rDPRP

rPLP-K I rPLP-D I rPLP-C rPLP-H rPL-IV I

rPLP-Cp I rPLP-N I rPLP-E I rPLP-F

rPLF-RP I rPLP-M I rPLP-A I

43129367 rpLp-L I

Fig. 1 Alignment of the rat PFU family locus

2 -

I

I

I I

I

4 - 5 -

Boxes with different sizes indicate different exons. Exon and intron lengths are only rough approximations.

. . . . I . . . . I .... I . . . . I . . . . I . . . . / . . . . ) . . . . I . . . . ( . . . . I . . . . I . . . . I . . . . I . . . . / . . . . I . . . . I

10 20 40 50 60 7 0 80

rPL-I MQLTLTLSGS GMQLLLLVSS LLLWENVAS K PTAIVSTDDL YHRLVEQSHN TFIMAADVYR EFDINFAKRS WMKDRILPLC 13~ rPL-IV MQLTLTLSGS ~QLLLLVSM LLLWEN@.S K ~ V S E E D L YHRLVEQSHN T ~ A A D V Y R EFDINFAKRS WMKDRILPLC

rPL-IV2 MQLTLTLSGS GMQLLLLVSS LLLWENVAS K PTAIVSTDDL YHRLVEQSHN TFIMAADVYR EFDINFAKRS WMKDRILPLC

. . . . l . . . . I . . . . [ . . . . I . . . . / . , . . I . . . . I . . . . [ . . . . / . . . . I . . . . I . . . . I . . . . l . . . . l . . . . l . . . . l

9 0 100 110 120 130 140 150 160

rPL-I HTASIHTPEN LEEVHEMKTE DFLNSIINVS VSWKEPLKHL VSAVTALPGA SVSMGKKAVD MKDKNLIILE GLQTLYNRTQ

rPL-IV HTASI&EN /&VH&TE &&&I& VSWKWIX&~ V S A V ~ P G A S!&&KAVD MKDKNLIILE G L Q ~ R T Q

rPL-IV2 HTASIHTPEN LEEVHEMKTE DFLNSIINVS VSWKEPLKHL VSAVTALPGA SVSMGKKAVD MKDKNLIILE GLQTLYNRTQ

. . . . l . . . . I . . .. I . . . . [ . . . . I . . . . I . . . . ) . . . . I . . . . I . . . . ( . . . . I . . . . I . . . . I . . . .

170 180 190 200 210 220

AKVEENFENF DYPAWSGLKD LQSSDEDTHL FAIYNLCRCF KRDIHKIDTY LKVLRCRWF KNECGVSTF rPL-I rPL-Iv rpL-Iv2

QWEE&NF DYPAWSGLKD LQSSDEDTHL FAIYNLCRCF ~ I H K I D T Y LKVLRCRVVF KNEC-----

AKVEENFENF @AWSGLKD ~SSDEDTHL FAIYNLCRCF KRDIHKIDTY LKVLRCRVVF KNECGVSTF

Fig. 2 Deduced amino acid sequences for rat PLI V2 The arrow indicates the predicted signd peptide cleavage site based on analyses using the Signalp glttp://www.cbs.dtu.dWserviceslSignalP/) soft-

ware. The outlined boxes show difFemnt residues compared with rPLZ gene.

considered, three rat PL- I genes, three mouse PL-Z genes and two deer mouse PL-Z genes clustered in a species-specific manner with high bootstrap supports

(loo%, 100% and 90%, respectively). A similar situa- tion was also observed for the PL-11 subfamily: in both mouse and rat only one copy of the PL-ZZ gene

594 Acta Genetica Sinica Vo1.33 No.7 2006

was identified, while three were identified in the deer mouse [16’ and they formed a distinct clade with 100% bootstrap support (Fig. 3). It is possible that PL-Z and PL-ZZ genes are species-specific and are products of

duplications that occurred after divergence of those species. But other evolutionary forces could also make paralogs within the same species more similar than orthologs from different species (see below).

2.3 Gene conversion test

Concerted evolution could possibly change the evolutionary record of a gene family. Gene conversion is known to be the most important mechanism of con-

certed evolution. To test the possible effect of this evolutionary force on rodent PL-I and PL-II subfami- lies, we used the method of Sawyer ‘17’, which is pro- posed to be a useful tool for detecting gene conver- sion events ‘**’. With the amino acid sequences of clas- sical members (including PRL, PL-I and PL-11) of the rodent PRL family as input file and with default setting of the program GENECONV 1.81 [lo], we did not de- tect any gene conversion cases among those two sub- families. Use of all PRL amino acid sequences or cDNA sequences as input file did not alter our result. This suggests that gene conversion cannot explain the species-specific phylogeny of PL-Z and PL-ZZ genes. In

100 c mGH mPLP-J rPLP-J mPLP-I rPLP-I gPL-I1 mPL-I1 $?L-I1 mPL-IIA

Ei$t dmPL-T1 mPL-T y mPL-T o mPL-T Q

rPL-TV2 rPLP-A mPLP-B rPLP-B mpLP-L rPLP-L Aocestor-PRL rPRL mPRL

mPLP-K

$HV mPLP-A

$E

$kF4 rPLP-K mPLP-M rPLP-M

$EF3 mPLP-N rPLP-N mPLP-0 mPLF-RP rPLF-RP mPLP-E r%YiiF %$W rDPW mPLP-C o kj&gcoy mPLP-C a rPLP-C rPLP-H

PL-I1 subfamaily

PL-I subfamaily

Fig. 3 Mouse and rat GH sequences are used as outgroups to locate the root of this tree. Bootstrap percentages (>50) by 500 replications are shown. Gene nomenclature follows that used in Wiemers et a1 16’. rn: mouse; r: rat; drn: deer mouse; g: golden hamster; J: Japa- nese vole.

MP tree based on amino acid sequences of all available rodent PRL genes

LI Ying et al.: Molecular Evolution of Prolactin Gene Family in Rodents 595

rPL- I V2 dmPL- I y

dmPL- I rPL- I1

mPL- I1 ~

addition, the relative chromosome locations of those PL-I genes differ between mouse and rat. In mouse, the three PL-I genes cluster together with similar length of intergenic sequences. However, rat PL-IV is not lo- cated close to the other two PL-I genes. Although the exact mechanism of this difference is not clear, the different chromosomal locations of PL-I between mouse and rat offer another evidence of independent gene duplication of PL-I genes in these two species. Thus it is evident that species-specific gene duplica- tions have occurred during evolution of PL-I and PL-11 genes in deer mouse, mouse and rat.

2.4 Estimation of gene duplication time

Since PL-I and PL-11 genes in rodent are derived from species-specific gene duplications, it becomes necessary to know when those duplications occurred. The time of divergence between mouse and rat is con- troversial, with studies showing it to be at 8-14 [lgl,

12-14 [201, 25 1211, 33 '221 or 41 Mya [231. Different di- vergence times will certainly give different gene du- plication times. Here we used the linearized tree method as implemented in MEGA 2.1 [13', to estimate those gene duplication times. We fust performed rela-

~ gPL- I1 -

tive rate test to evaluate the molecular clock hypothesis

of synonymous substitutions among PRL, PL-I and

PL-11 genes using RRTree ["I and found no rate varia- tion, which suggested that the molecular clock hy- pothesis held here. Then we constructed neighbor- joining(NJ)tree of PL-I and PL-11 genes in rodents with PRL from mouse and rat as the outgroups to locate the root of the tree. Using the average ds rate (5.3 x loT9 synonymous substitution/site/year) estimated from 1 1 000 orthologous genes between mouse and rat [243 251, as the evolutionary rate in a linearized NJ tree, we esti- mated that gene duplications leading to PL-I genes in mouse, rat and deer mouse started at 3.78, 9.58 and 10.14 Mya, respectively, and that deer mouse PL-I1 gene duplication started at 5.66 Mya (Fig. 4). Our results suggest that those gene duplications leading to a cluster of PL-I and PL-I1 in rodents occurred very recently.

3 Discussion

To date, 26 mouse [61 and 21 rat PRL members have been identified by utilizing the published ge- nomic information, but only pituitary PRL gene exists

Fig. 4 Linearized NJ tree for classic PRL members ds distances were used.

596 Acta Genetica Sinica Vo1.33 No.7 2006

in human, chimpanzee and dog genomes. Although a cluster of PRL-like genes also exists in ruminants, phylogenetic analysis suggests that these genes in rodent and in ruminant evolved independently [261.

Additionally, it has been suggested that glires is a sister taxon to primate [I4] . Thus the expansion of PRL

gene family identified in mouse and rat is specific to rodents and the gene duplications leading to this fam- ily occurred after the divergence of rodents from other mammals.

We note that gene duplications occurred continu- ally to produce members with different exodintron structure after the appearance of the extra exon (Fig. 3). It has even continued after the divergence of deer mouse, mouse and rat. Vrana et al. 'I6] believed that the PL-I and PL-II genes in deer mouse are spe- cies-specific. After adding more family members, we can draw the conclusion that PL-I and PL-I1 sub- families in mouse and rat are also species-specific. A further analysis shows that these species-specific gene duplications occurred very recently. Additionally, although the three mouse PL-I genes share more than 98% amino-acids sequence similarity, they are ex- pressed at different times of gestation and in different cell types [@. Similar temporal and/or cell specificity have also been described for rat and deer mouse PL-I genes [16. 271 . This suggests that these PL-I genes, which originated after the divergence of the three species, have become functionally divergent. And thus the origins of these new genes are possible ad- aptations to more specific functional allocation. But the evolutionary significance of these species-specific gene duplications is not clear, because of the lack of experiments comparing the functional and expres- sional differences among different PL-I genes in dif- ferent species. The existence of different exodintron structure and species-specific gene cluster suggests that multiple gene duplications and diversification have occurred during the evolution of this gene fam- ily. Thus frequent gene duplication and fast diver- gence may be the main evolutionary mechanism of this gene family in rodent.

Acknowledgments: We thank Prof. Michael Wallis

from University of Sussex for his valuable sugges- tions and comments.

References:

191

Goffin V, Shiverick K T, Kelly P A, Martial J A. Se- quence-function relationships within the expanding fam- ily of prolactin, growth hormone, placental lactogen, and related protein in mammals. Endocrine Reviews, 1996, 17 : 385-410.

Forsyth I A, Wallis M. Growth hormone and prolactin- molecular and functional evolution. Journal of Mummaly

Gland Biology and Neoplusiu, 2002, 7(3) : 291-312.

Wallis M. The expanding growth hormone/prolactin fam- ily. Journal of Molecular Endocrinology, 1992, 9 :

185-1 88. Soares M J, Miller H, Orwig K E, Peters T J, Dai G. The uteroplacental prolactin family and pregnancy. Biology of

Reproduction, 1998, 58 : 273-284. Jackson-Grusby L, Pravtcheva D, Ruddle F H, Linzer D I H. Chromosomal mapping of the prolactidgrowth hor- mone gene family in the mouse. Endocrinology, 1988, 122 : 2464-2466.

Wiemers D 0, Shao L J, Ain R, Dai G Soares M J. The mouse prolactin gene family locus. Endocrinology, 2003, 144(1) : 313-325.

Cooke N E, Szpirer C, Levan G. The related genes en- coding growth hormone and prolactin have been dis- persed to chromosomes 10 and 17 in rat. Endocrinology,

1986, 119 : 2451-2454.

Jeanmougin F, Thompson J D, Gouy M, Higgins D G, Gibson T J. Multiple sequence alignment with Clustal X.

Trends in Biochemical Sciences, 1998,23 : 403-405.

Swofford D L. PAUP*: phylogenetic analysis using par-

simony (* and other methods). Version 4.0 Sinauer

Associates, Sunderland, Mass. 1998. [lo] Sawyer S A. GENECONV: statistical tests for detecting

gene conversion-version 1.81. Department of Mathe- matics, Washington University, St. Louis, Mo. 2000.

[ 111 Li P, Bousquet J. Relative-rate test for nucleotide substi- tutions between two lineages. Molecular Biology and

Evolution, 1992,9 : 1185-1189.

[12] Rechavi R M, Huchon D. RRTree: relative-rate tests be- tween groups of sequences on a phylogenetic tree. Bioin- formatics, 2000, 16 : 296-297.

[I31 Kumar S, Tamura K, Jakobsen I B, Nei M. MEGA2: molecular evolutionary genetics analysis software. Bio-

informatics, 2001, 17 : 1244-1245. [I41 Murphy W J, Eizirik E, Johnson W E, Zhang Y P, Ryder

0 A, O'Brien S J. Molecular phylogenetics and the ori-

LI Ying et al.: Molecular Evolution of Prolactin Gene Family in Rodents 597

gins of placental mammals. Nature, 2001,409 : 614-618. [I51 Wallis M. Episodic evolution of protein hormones: mo-

lecular evolution of pituitary prolactin. J o u m l of Mo-

lecular Evolution, 2000, 50 : 465-473. [16] Vrana P B, Matteson P G, Schmidt J V, Ingram R S, Joyce

A, Prince K L, Dewey M J, Tilghman S M. Genomic im- printing of a placental lactogen gene in Peromycus. De-

velopment Genes and Evolution, 2001,211 : 523-532. [17] Sawyer S. Statistical tests for detecting gene conversion.

Mol Biol Evol, 1989,6 : 526-538. [18] Drouin G, Prat F, Ell M, Clarke G D. Detecting and char-

acterizing gene conversions between multigene family members. Molecular Biology and Evolution, 1999, 16

(10) : 1369-1390. [19] Jacobs L L, Pilbeam D. Of mice and men: fossilbased

divergence dates and molecular “clocks”. Journal of Human Evolution, 1980, 9:551-555.

[20] Robinson M, Catzeflis F, Briolay J, Mouchiroud D. Mo- lecular phylogeny of rodents, with special emphasis Murids: evidence from nuclear gene LCAT. Molecular

Phylogenetics and Evolution, 1997, 8 : 423-434.

[21] O’huigin C, Li W H. The molecular clock ticks regularly

in muroid rodents and hamsters. Journal of Molecular

Evolution, 1992,35 : 377-384. [22] Nei M, Xu P, Glazko G. Estimation of divergence times

from multiprotein sequences for a few mammalian spe- cies and several distantly related organisms. PNAS, 2001, 98 : 2497-2502.

[23] Kumar S, Hedges S B. A molecular timescale for verte- brate evolution. Nature, 1998, 392 : 917-920.

[24] Rat Genome Sequencing Consortium. Genome sequence of the brown Norway rat yields insights into mammalian evolution. Nature, 2004,428 : 493-521.

[25] Grus W E, Bang J Z. Rapid turnover and species-specificity of vomeronasal pheromone receptor genes in mice and rats. Gene, 2004,340 : 303-312.

[26] Hiraoka Y, Ogawa M, Sakai Y, Takeuchi Y, Komatsu N, Shiozawa M, Tanabe K, Aiso S. PLP-I: a novel prolac- tin-like gene in rodents. Biochimica Biophysica Acta, 1999, 1447 : 291-297.

[27] Nieder G L, Jennes L. Production of placental lactogen-I by trophoblast giant cells in utero and in vitro. Endocri-

nology, 1990,126 : 2809-2814.