AtUCP2: a Novel Isoform of the Mitochondrial Uncoupling Protein of ...

Plant Cell Physiol. 40(11): 1160-1166 (1999) JSPP © 1999

AtUCP2: a Novel Isoform of the Mitochondrial Uncoupling Protein of Arabidopsis thaliana Akio Watanabe, Mikio Nakazono, Nobuhiro Tsutsumi and Atsushi Hirai 1 Laboratory of Plant Molecular Genetics, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo, 113-8657 Japan

Key words: Arabidopsis — Gene expression — Low temperature — Mitochondria — Uncoupling protein.

Certain types of plants generate heat during flowering (for review, see Seymour 1997). However, relatively little is known at present about the molecular mechanism enabling the heat generation in higher plants. In mammals, mitochondrial uncoupling proteins (UCPs) play a central role in the heat generation in response to low temperature (termed adaptive thermogenesis). These proteins, encoded in the nuclear genome, contain six transmembrane domains and are imported into the mitochondrial inner membrane without the help of any transit sequences (Liu et al. 1988). In mitochondria, they dissipate the proton gradient formed through respiration without the synthesis of ATP. The resulting freed energy is readily converted to heat, helping Abbreviations: BAT, brown adipose tissue; EST, expressed sequence tag; ORF, open reading frame; UCP, uncoupling protein. The nucleotide sequence reported in this paper has been submitted to DDBJ under accession number AB021706. 1 Corresponding author.

the animals to maintain their body temperature in cold environments. One of the major expression sites of the proteins, brown adipose tissue (BAT), is known as a center of heat generation during cold adaptation in rodents (Nicholls and Locke 1984). Recently, a biochemical approach toward the dissection of plant mitochondria has revealed an uncoupling protein functioning in potato mitochondria (Vercesi et al. 1995). Subsequently, a potato cDNA for this protein (named StUCP) was isolated (Laloi et al. 1997). StUCP was shown to possess all the typical features reported for mammalian UCPs, and also to function as an uncoupling protein when overexpressed in yeast (Laloi et al. 1997). It has also been shown that cold treatment enhances expression of the gene in the plant. In mammals, the UCP gene constitutes a small multigene family consisting of at least three members (Boss et al. 1997a, b). These UCP gene members have different expression sites and different responses to low temperature. In rodents, the UCP1 gene is expressed mainly in BAT, and cold treatment enhances the accumulation of the transcripts (for a review, see Ricquier et al. 1991). UCP2 mRNA is found in most types of tissues of rodents, as well as humans (Boss et al. 1997a, b), while human UCP3 transcripts accumulate specifically in skeletal muscle (Boss et al. 1997a). Unlike the occurrence of the multigene family for mammalian UCPs, only a single species of UCP cDNA has been isolated in potato so far. Moreover, the latest study, published in the midst of our study, has suggested that Arabidopsis has only a single gene to encode its UCP (Maia et al. 1998). In the present study, we found a novel UCP gene in the Arabidopsis genome, demonstrating for the first time in higher plants the occurrence of multiple genes for the protein. A Northern blotting analysis suggested that low temperature does not directly induce the expression of the novel UCP gene of Arabidopsis. Materials and Methods Isolation of a genomic clone for an Arabidopsis UCP—To amplify a genomic fragment encoding Arabidopsis UCP, we synthesized the TM3 and the TM6 degenerative primers, based on the amino acid sequences found in the third and the sixth transmembrane domains of potato UCP (ITIANPTDL and GSWNVIMF, respectively). PCR conditions consisted of preheating at 94°C for nine min, 30 cycles of one min at 94°C, two min at 45°C, and two

1160

Downloaded from http://pcp.oxfordjournals.org/ by guest on October 26, 2016

Mitochondrial uncoupling proteins (UCPs) play a central role in adaptive thermogenesis in mammals. The UCPs dissipate the proton gradient formed through respiration without ATP synthesis, and the freed energy is readily converted to heat, helping the animals to maintain their body temperature in cold environments. Recently, it was found that UCPs also function in plant mitochondria. Subsequently, a cDNA clone encoding a UCP in potato was isolated. Whereas the UCP gene constitutes a multigene family in mammals, only a single cDNA sequence has been reported so far for the potato UCP. Moreover, it has been recently suggested that Arabidopsis has only a single nuclear gene for UCP. Here we report the existence of another UCP gene in the Arabidopsis genome, showing for the first time the occurrence of a multigene family for the protein in higher plants. A cDNA analysis of this gene showed that the novel isoform possesses all typical features reported for known UCPs. However, the new gene, unlike the other gene in Arabidopsis and the gene in potato, did not appear to respond to low temperature.

Novel isoform of Arabidopsis uncoupling protein

(underline shows a BamHl site introduced in the sequence) SP2: GAAAGATTCCCACTGGAGATGGTGAGA SP3: GACTGTGTTTCGGTAAGTAGAG SP3C: complimentary to SP3 Determination of nucleotide sequence—The nucleotide sequence was determined using an automatic DNA sequencer model 373S (Perkin Elmer ABD). In the analysis of the PCR products, we sequenced several independent clones in order to eliminate the variance introduced by the DNA polymerase. Amino acid sequence alignment of AtUCP2 with other known UCPs—AtUCP2 was aligned with other known UCPs using the CLUSTAL W algorithm (Thompson et al. 1994). For shading residues in the aligned sequences, we used the BOXSHADE program (ISRC Bioinformatics Group). Genomic Southern blotting analysis—In the Southern blotting analysis, three /jg of Arabidopsis DNA were digested with either EcoRl or Xbal. The restricted DNA was separated on an agarose gel, and blotted onto a nylon membrane. Either the AtUCPl or the AtUCP2 cDNA fragment encompassing the Cterminal and the 3' untranslated regions was labeled with a DIG DNA Labeling and Detection Kit (Roche Diagnostics GmbH, Swiss). These labeled fragments were used as gene-specific probes for hybridization. Hybridization was carried out as described previously (Ito et al. 1997). To detect the fragments related to

UCP genes, another probe was prepared by amplifying and labeling an AtUCP2 cDNA spanning the region between the SP1 sequence and the C terminus of the ORF. Northern blotting analysis—The response of the UCP genes to low temperature was tested using two groups of Arabidopsis seedlings. First, a group of the plants (Columbia, wild type) was grown for four weeks on vermiculite beds at 24°C under continuous light, and then transferred into a cold room set at 4°C. After 24, 48 and 72 h of incubation, total RNA was extracted from the plants. During the cold treatment, the plants were kept under weak light to avoid dark-induced yellowing of the leaves. We also grew another group of plants (Columbia gll). These seedlings were planted on MS medium containing MS salts (Murashige and Skoog 1962) and 2% sucrose at 24°C under continuous light. Two weeks after germination, the seedlings were cold-treated at 4°C in darkness. Total RNA was prepared from seedlings that had been cold-treated for 0, 24 and 48 h. Hybridization was carried out as described previously (Ito et al. 1997), using the gene-specific probe for either AtUCPl or AtUCP2. Results and Discussion Isolation of a genomic clone for an Arabidopsis UCP, AtUCPl, and its gene structure—When we started this study, no nucleotide sequence had been reported for Arabidopsis UCPs. Although a few expressed sequence tag (EST) clones showed homology to known UCP genes, it was unclear where and when such a possible UCP gene was expressed in the plants. So, we started our study by isolating the genomic clone for the gene, aiming to first clarify the whole ORF sequence. To obtain a probe for screening, we carried out PCR as described above, with Arabidopsis DNA as a template. Also, taking into account the possibility of multiple genes, we used degenerative primers that covered the regions conserved among various UCPs. In this experiment, we could obtain only a single DNA fragment, even though degenerative primers were used in the reaction. However, sequence analysis of the amplified fragment strongly suggested that it was a part of a gene for an Arabidopsis UCP (data not shown). With this fragment as a probe, we screened a library, and isolated several genomic clones for the gene. At this time, Maia et al. (1998) reported a complete mRNA sequence for an Arabidopsis UCP, and they referred to the protein as AtPUMP for a plant uncoupling mitochondrial protein (EMBL accession no. AJ223983). We then knew that the gene we isolated encoded their AtPUMP. In this report, though, we designate this protein as AtUCPl for convenience, since we describe in the following sections a novel isoform of Arabidopsis UCP, AtUCP2. Comparison of the determined genomic sequence with the AtUCPl mRNA sequence revealed the whole gene structure of the AtUCPl gene, as shown in Fig. 1. In most mammalian UCP genes, the sequence coding the protein is found to be separated into six regions, each of which contains a transmembrane domain (Kozak et al. 1988, Cassard et al. 1990). The figure also depicts the gene structure of mouse UCP1 (Kozak et


min at 72°C, and final extension at 72°C for nine and a half min. The sequences of the primers were: TM3: T(T/C/A)AC(A/T)AT(A/T)GCAAATCC(CVT)AC(A/T)GATCT TM6: AAACAT(A/G)AT(T/G)AC(A/G)TTCCA(A/T)GAGCC Amplification of cDNA fragments encoding a novel UCP isoform, AtUCP2—For the amplification of AtUCP2 cDNA fragments, total RNA was extracted from 14-day-old seedlings of Arabidopsis thaliana (Columbia gll) kept at 4°C in darkness for 48 h. In this experiment the seedlings were grown on MS medium containing MS salts (Murashige and Skoog 1962) and 2% sucrose at 24°C under continuous light. They were subjected to the cold treatment two weeks after germination. First strand cDNA was synthesized from three ptg of total RNA extracted from the seedlings, using an oligo dT primer and Superscript II RT (GIBCO BRL). With this cDNA as a template, we amplified three cDNA fragments, which together covered the whole open reading frame (ORF) of AtUCP2. All the primers used in the reaction (SP1 to SP3 and SP3C, see the underlined sequences in Fig. 2) were designed based on the sequence on chromosome 5 deposited in the database (DDBJ accession no. AB016885). First, a fragment specifying the central portion of the protein (the SP2-SP3 region) was amplified by PCR using the SP2 and the SP3 primers. Next, a fragment covering the C-terminal portion (the region downstream of SP3) was obtained similarly through two rounds of PCR (using the SP2 primer and an oligo dT primer first, and then the SP3C and the oligo dT primers). Finally, the N-terminal portion (SP1SP3) was amplified as follows. Since we found in the genome sequence a perfect consensus sequence around the translation initiation sites (AACAATGGC, Liitcke et al. 1987), we considered this ATG as a putative start site of the ORF. Using the SP1 primer covering this site and the SP3 primer, the cDNA fragment of the N-terminal portion was amplified. In all PCR experiments, AmpliTaq GOLD (Perkin Elmer CETUS) was used. The PCR conditions consisted of preheating for nine min at 94°C, 35 cycles of 30 s at 94°C, 30 s at 54°C and one min at 72°C, and final extension at 72°C for nine min. The primers used in this experiment were: SP1: GGGATCCTATAGCATAACAATGGCGGAT

1161


1162 IV

V

VI

TCCTATAGCAIA>CAATGGCCG*TTTCMACCMGGATCCAGATITCtiTTCCTIGAAACCTTCAITTCCAGCGCrTIC

Mouse UCP1

""'

H

A

D

F

K

P

R

I

E

I

S

F

L

E

T

F

I

C

S

A

78

F

GCTGCTTGTTTTGCIGAGTTATGTACTATACCGTTGGACACAGCCAAAGTTAGACTTCAGCTTCAAAGAAAIiATTCCC 156 A

IV

V

VI

100 bp

T

AtUCPl

A

C

F

A

E

L

C

T

I

P

L

O

T

A

K

V

R

L

O

I

Q

R

K

I

1

P

0

*

'

ACTGGAGATGGTGAGAATTTGCCCAAGTATAGAGGArCAATTGOIACTCTAGCTACCATAGCIAGAGAAGAAGGTATT 2 3 4 G

0

G

E

N

L

P

K

Y

R

G

S

I

G T L A T

I

A R

E

£

G

I

rCAGGTCTTTGGAAAGGTGrTATTGCAGGACTTCAICGCCAATGTATCTATGGTGCCTTAAGGATTGGGTTATATGAG 3 1 2 S

G

L

1

K

G

V

I

A

G

L

H

R

0

C

I

Y

G

G

I

R

I

G

L

Y

E

CCTGTGAAGACACTTITGOnGGAAGTGACTIIAITGGCGATATTCCTTTAIATCAAAAGAIICTTGCAGCTTTGTTA 3 9 0 P

AWCP2

V

K

T

L

L

V

G

S

D

F

I

G

O

I

P

L

Y

O

K

I

L

A

A

L

L

ACTGGAGCTATAGCTATTATTGTAGCTAAICCAACTGATCTTGTTAAAGTTCGGCTTCAATCAGAAGGAAAGTTACCG 4 6 8 T G A I A I

I

V

A

H

P

I

D

L

V

K

V

R

L

O

S

E

G

K

L

P

GCTGGGGTICCTAGGCGTTATGCAGGAGCTGTAGACGCTIAITTCACCATTGIGAAGCIGGAAGGAGTTAGTGCGCIA 5 4 6 A

G

V

P

R

R

r

A

G

A

V

D

A

T

F

T

I

V

K

L

E

G

V

S

A

L

TGGACTGGACITGGTCCCAATATTGCCCGGAATGCTATTGTAAATGCTGCAGAGTTAGCIAGTTATGATCAAATAAAG 6 2 4 •

T

G

L

G

P

N

I

A

R

N

A

I

V

N

A

A

E

L

A

S

Y

O

O

I

K

GAGACAATTATGAAAATTCCGTTCTTCAGAGACAGTGTTTTAACTCATCIACTAGCTGGTTTAGCTGCAGGCTTCTTC 7 0 2 E

T

I

U

K

I

P

F

F

R

D

S

V

L

T

H

L

L

A

G

I

A

A

G

F

F

GCTGrCTGCATCGGITCTCCAATTGATGTGGTGAAAICTAGAATGATGGGAGACTCrACTTACCGAAACACAGTCGAT A

V

C

I

G

S

P

I

O

V

V

K

S

R

H

H

O

D

S

T

Y

R

H

T

V

O

"

"

780 1

rGCTTCATCAAAACGAIGAAGACCGAGGGGATIATGGCAITCTACAAAGGATTTCTCCCGAATITTACACGGCTAGGA 8 5 8 C F I K T U K T E G I I I A F T K G F .

al. 1988) for comparison. Unlike the mammalian gene, the sequence encoding AtUCPl was found separated into as much as nine short regions. Thus, we could not find a correspondence between the respective coding regions and the transmembrane domains in the plant gene. On the other hand, we noted that excision of all the introns followed the GU/AG rule (Brown et al. 1996) in this gene (data not shown). A novel gene for a UCP isoform of Arabidopsis— According to Maia et al. (1998), Arabidopsis has only a single gene to encode its UCP. In agreement with this notion, we could amplify only a single DNA fragment from Arabidopsis DNA in the previous experiment. However, as a result of a database search, based on the AtUCPl genomic and mRNA sequences, we found another possible UCP gene in a partial sequence deposited for chromosome 5 of Arabidopsis (77.8 kb, DDBJ, accession no. AB016885). The nucleotide sequence of the possible UCP gene was apparently different from the sequence we determined for the AtUCPl gene (the AtUCPl gene was not located in the 77.8 kb region of chromosome 5), but the predicted gene product showed high homology with AtUCPl and other known UCPs on an amino acid sequence basis. Considering that the nucleotide sequence for the sixth transmembrane domain of the product differs slightly from that of AtUCPl (the domain structure of AtUCP2 is also shown by bold lines (I to VI) in Fig. 1), we assume that this difference prevented the TM6 primer used in the previous PCR experiment from annealing to the gene on the chromosome. Isolation of cDNA fragments coding AtUCP2, an isoform of Arabidopsis UCP—To determine whether the sequence on chromosome 5 is actually transcribed in the plants, we tried to amplify the cDNA fragments originating from its transcripts. As a template for the amplification, we synthesized 1st strand cDNA from the total RNA extract-

L P N F T R L G

ACCTGGAATGCCATTATGTTCCTCACATTAGAACAAGTGAAAAAAGTGTTTCTAAGAGAAGTCTTGTACGATTGATTC 9 3 6 T

»

N

A

I

»

F

L

T

I

E

O

V

K

K

V

F

L

R

E

V

L

Y

O

«

TCAGATCCCIAGTCGAAAACCATACCTTATTACAIAATCCCTTCTATAAAACTTTGAATTGTTAGAATTAAAACATAT 1014

ATACTTTCTATGTJt]»TGTGAGCTTTGTTATTTAGATTAGTATAGAAACATTTTATCCAAAAAAAAAITCTTTG§AAGA 1092

ACAGTTTATACGAACGGTCTCTTTATTT§

p o l y (A)

Fig. 2 Nucleotide sequence of the ORF encoding a novel UCP isoform of Arabidopsis, AtUCP2. The regions to which the specific primers (SP1 to SP3) anneal are underlined. The amino acids deduced from the cDNA sequence are given as single letters. The asterisk shows a stop codon. Nucleotides shown in boxes are multiple polyadenylation sites.

ed from the plants kept at 4°C for 48 h. As described in Materials and Methods, we obtained several partial cDNA fragments which, together, covered the whole ORF. Fig. 2 shows the ORF and its amino acid sequence. The protein encoded by the ORF consists of 305 amino acids (which is almost the same number as in other plant UCPs), and exhibited 85% homology with both AtUCPl and the potato UCP, and around 60% homology with other mammalian UCPs. We therefore concluded that the sequence found on chromosome 5 is actually functioning as a gene for a novel UCP isoform. We designated here the protein encoded by the gene as AtUCP2. An alignment of AtUCP2 with other UCPs in Fig. 3 showed that AtUCP2 possesses typical features of UCPs, i.e. six transmembrane domains, the typical mitochondrial carrier signatures, and a possible nucleotide binding region (Bouillaud et al. 1994). Also, the N-terminal portion of the protein shares high homology with the corresponding parts of AtUCPl and StUCP. This makes it more likely that the translation initiation site that we tentatively identified is correct. Genomic organization of the AtUCP2 gene—Subsequently, we clarified the genomic organization of the gene for AtUCP2 by aligning its cDNA sequence with the ge-


Fig. 1 Comparison of the gene structures of AtUCPl and AtUCP2 of Arabidopsis. For comparison, the structure of Mouse UCP1 gene is shown in parentheses. Open boxes with Arabic numerals show the regions encoding the UCP ORFs. The bold lines above the boxes show respective transmembrane domains (I to VI).


1163 TH 2

AtUCP2 AtUCPI StUCP

1 —tMFKPRIEISFLETfllCS,

NBS

Fig. 3 Comparison of AtUCP2 with other known UCPs. Identical amino acids are shown in black, and similar ones in outlined letters. The regions with single broken lines correspond to transmembrane domains (from TMl to TM6). The residues with a double underline are a predicted nucleotide binding site (NBS). The vertical lines connect the typical mitochondrial carrier signature found in each repeating unit (PxD/ExxKxR-(20-30 residues)-D/EG-(four amino acids)-(an aromatic amino acid)-KG). Additional gaps were introduced, besides the ones that the program did, into the alignment in order to place respective repeating units in parallel. All the amino acid sequences of known UCPs were obtained from the databases. Respective accession numbers are as follows: StUCP (potato UCP, EMBL Y11220); HsUCPl (human UCP1, GenBank U28480); HsUCP2 (human UCP2, GenBank U82819); HsUCP3 (human UCP3, GenBank NM0O3356); MmUCPI (mouse UCP1, GenBank U63419); MmUCP2 (mouse UCP2, GenBank U69135); MmUCP3 (mouse UCP3, GenBank AF032902).

nomic sequence of chromosome 5. In Fig. 1, the genomic organization of the AtUCP2 gene is aligned with that of the AtUCPI gene. The sequence encoding AtUCP2 was found to be separated into nine regions, as was found previously in the AtUCPI gene. So, unlike mammalian UCP genes, we could not find a correspondence between

the respective coding regions and the transmembrane domains in this gene, as well as in the AtUCPI gene. Interestingly, the ORFs for both AtUCPs were interrupted by introns at exactly the same sites with respect to their domain structures. We speculate that such fragmentary structures are specific to plant UCP genes. A variety of


HsUCPl MmUCPI HsUCP2 MmUCP2 HsUCP3 IMJCP3

1164


ting analysis. As an /ifL/C/^-specific probe, we labeled an AtUCP2 cDNA fragment which covered a C-terminal portion of the ORF and the 3' untranslated region (shown by a striped box in Fig. 5a). The specificity of the probe was confirmed by a Southern blotting analysis shown in Fig. 5b. As the blot shown by "AtUCP2" in the panel (b) demonstrates, the probe hybridized specifically to the fragments corresponding to the AtUCP2 gene (labeled "2"), but not to the fragments derived from the AtUCPl gene (labeled " 1 " in the blot on the left). In the panel (c), a blot was presented, for comparison, that was hybridized with another cDNA probe covering the whole AtUCP2 ORF (shown by an open box in Fig. 5a). As the panel (c) shows, this cDNA probe detected intense signals corresponding to the AtUCP2 gene fragments (labeled "2", note that this

(a)

A1UCP2 cDNA probes

X I '

nnn AlVCPl cDNA probe

(b)

i.o kb

(c) AtUCPl BUD EX

AtUCP2 M X

kb

— 23.1 —

kb

AtVCP2 E

X

•

23.1-

— 9.4 — — 6.6

-

MmUCP3

^m

— 4.4 —

— 2.3 — — 2.0 —

ABJCP2

Fig. 4 Phylogenetic relationships among plant and mammalian UCPs. A phylogenetic tree was drawn using TREEVIEW (Page 1996). The analyzed UCPs include BtUCPl (bovine UCP1, EMBL X14064), SsUCP2 (pig UCP2, GenBank AF036757) and BtUCP3 (bovine UCP3, GenBank AF092048), besides those appearing in Fig. 3. The number at each node shows the reliability value of the branch, which was calculated from 1000 quartet puzzling steps. The scale bar represents 0.1 mutations/site.

Fig. 5 (a) Restriction maps of the AtUCP genes, and the cDNA probes used in the analysis. Black boxes show the regions coding AtUCP2 (top) or AtUCPl (bottom). Striped boxes show the AtUCP2-specific and the /U£/CP/-specific probes, respectively. The AtUCP2 probe covering the who\e AtUCP2 ORF is shown by an open box. E, EcoRl, X, Xbal. (b) Genomic Southern blotting analysis. Genomic DNA of Arabidopsis, digested with £coRI or Xbal, was separated on an Agarose gel and transferred onto a nylon membrane. The /Ut/C/2-specific probe hybridized to an EcoRl fragment (about 8 kb) and a Xbal fragment (about 7 kb) (labeled "2" in the blot in the center), whereas the AtUCPl-specific probe detected an EcoRl fragment (about 5 kb) and two Xbal fragments (about 3.5 kb and 1.8 kb) (labeled " 1" in the blot on the left), (c) The AtUCP2 cDNA probe covering the whole AtUCP2 ORF produced several extra signals (shown by open triangles), other than the signals corresponding to the AtUCP2 gene or the AtUCPl gene (labeled " 1 " or "2").


polyadenylation sites was also found in both genes. The boxed letters in Fig. 2 indicate the multiple polyadenylation sites found in AtUCP2 cDNAs. In humans, short and long UCP3 proteins are generated from a common UCP3 gene depending on the transcription termination sites (Solanes et al. 1997). However, the variety of polyadenylation sites did not seem to affect the C-terminal sequence of either of the Arabidopsis UCPs. Phylogenetic relationship of plant UCPs with other known UCPs—The phylogenetic tree in Fig. 4 provides another insight into the plant UCPs. According to the program used in this study, all of the three known plant UCPs were classified into a single isolated group, which did not belong to any of the three subclasses of mammalian UCPs. It was also shown that among the plant UCPs, the potato UCP, StUCP, is more closely related to AtUCPl than to AtUCP2. This result may raise the possibility of the existence of another potato UCP isoform which is equivalent to AtUCP2. It is of interest to know whether the UCP gene constitutes a multigene family in potato. Low temperature does not induce the expression of the AtUCP2 gene—Cold-enhanced accumulation of the transcripts for the potato UCP gene has been demonstrated (Laloi et al. 1997). Moreover, in Arabidopsis, 48 h of incubation at 4°C increased the transcript level of AtUCPl to a peak amount (Maia et al. 1998). To determine the response of the novel UCP gene of Arabidopsis to low temperature, we subsequently carried out a Northern blot-

Novel isoform of Arabidopsis uncoupling protein probe also covers a 6 kb Xba\ fragment), as well as other weak signals. While the signals labeled " 1 " seemed to correspond to the AtUCPl gene judging from their sizes, the rest of them with open triangles may suggest the existence of other members of the UCP gene family of the plants. However, the possibility can not be excluded that these signals are derived from genes for other UCP-related proteins. Because of the similarity to other mitochondrial translocators, UCPs have been classified as members of the mitochondrial carrier family (Klingenberg 1990).

(a)

4°C

u AtUCP2 AtUCPl

v

Weak light 24

*..—

R.T.

48 72 • • •

Dark

72h

4°C 0

24 48h

-

—

•

25S rRNA Fig. 6 Response of the AtUCP2 gene to low temperature, (a) Arabidopsis seedlings (grown on vermiculite for four weeks) were incubated at 4°C for 0, 24, 48 and 72 h in weak light. Total RNA was extracted from the seedlings and also from the control plants kept at room temperature (24°C) for 72 h in light. Twenty p% of the RNA were loaded on each lane. Hybridization was carried out, using the AtUCP2-specific probe or, for comparison, the /4ft/CiJ/-specific probe. In the cold-treated plants, no significant increase was observed in the level of the AtUCP2 transcripts, and also in the transcript level of AtUCPl, which had been shown to be cold inducible by Maia et al. (1998). (b) A different group of seedlings (14-day-old seedlings grown on MS medium) was subjected to cold treatment in darkness for 0, 24 and 48 h. Fifteen jug of the RNA were loaded on each lane. Whereas the amount of the AtUCP2 transcripts remained at low levels during the treatment, the level of the A tUCPl transcripts increased significantly after 48 h of the treatment, in good agreement with the report by Maia et al. (1998).

at present. We assume, though, that differences in the age of seedlings, plant growth conditions, the procedure for cold treatment etc., could be responsible for the different responses of the AtUCPl gene to low temperature. Evidence supporting our hypothesis came from the following experiment. In the subsequent experiment, we grew Arabidopsis seedlings on MS medium for two weeks, and incubated them at 4°C in darkness rather than in weak light as in the previous experiment. In these seedlings, the amount of the AtUCPl transcripts increased after 48 h of the cold treatment (Fig. 6b), in good agreement with the report by Maia et al. (1998). From these observations, it is suggested that factors other than temperature can affect the transcript level of the AtUCPl gene in the cold-treated plants. Unlike the AtUCPl gene, no significant increase in the amount of the AtUCP2 transcripts was observed also in darkness (Fig. 6b). Thus, this novel UCP gene did not seem to be cold-inducible. In summary, we demonstrated in this study the existence of a novel UCP gene of Arabidopsis, showing for the first time the occurrence of a multigene family for plant UCPs. Unlike the AtUCPl gene, no significant enhancement by low temperature was observed for the expression of the AtUCP2 gene in our experiments. This observation may suggest that these two UCP genes are differentially regulated under cold environments. We hope that our study will facilitate further studies on UCPs of higher plants. This work was supported partly by grants-in aid from the Ministry of Education, Science and Culture of Japan and by grants from the Program for Basic Research Activities for Innovative Biosciences (PROBRAIN). The authors thank D. Saisho of our laboratory for preparing the RNA samples. References Boss, O., Samec, S., Dulloo, A., Seydoux, J., Muzzini P. and Giacobino, J.-P. (1997b) Tissue-dependent upregulation of rat uncoupling protein2 expression in response to fasting or cold. FEBS Lett. 412: 111-114. Boss, O., Samec, S., Paoloni-Giacobino, A., Rossier, C , Dulloo, A., Seydoux, J., Muzzin, P. and Giacobino, J-P. (1997a) Uncoupling protein-3: a new member of the mitochondrial carrier family with tissuespecific expression. FEBS Lett. 408: 39-42. Bouillaud, F., Arechaga, I., Petit, P.X., Raimbault, S., Levi-Meyrueis, C , Casteilla, L., Laurent, M., Rial, E. and Ricquier, D. (1994) A sequence related to a DNA recognition element is essential for the inhibition by nucleotides of proton transport through the mitochondrial uncoupling protein. EMBO J. 13: 1990-1997. Brown, J.W.S., Smith, P. and Simpson, C.G. (1996) Arabidopsis consensus intron sequences. Plant Mol. Biol. 32: 531-535. Cassard, A.-M., Bouillaud, F., Mattei, M.-G., Hentz, E., Raimbault, S., Thomas, M. and Ricquier, D. (1990) Human uncoupling protein gene: structure, comparison with rat gene, and assignment to the long arm of chromosome 4. J. Cell. Biochem. 43: 255-264. Ito, Y., Saisho, D., Nakazono, M., Tsutsumi, N. and Hirai, A. (1997) Transcript levels of tandem-arranged alternative oxidase genes in rice are increased by low temperature. Gene 203: 121-129. Klingenberg, M. (1990) Mechanism and evolution of the uncoupling protein of brown adipose tissue. Trends Biol. Sci. 15: 108-112.


Using the AtUCP2-specific probe, we tested the response of the gene to low temperature. In this experiment, Arabidopsis seedlings were cold-treated in weak light. As the blot in Fig. 6a shows, the transcript level remained at low levels during the treatment, suggesting that the cold treatment does not induce the gene directly. We also examined the amount of AtUCPl transcripts in the coldtreated seedlings for comparison, since Maia et al. (1998) reported that 48 h of incubation at 4°C increased the transcript level of AtUCPl to a peak amount. The transcript level, however, even decreased after 48 h of cold treatment in our experiment, although it recovered slightly after 72 h of the treatment. The recovered level, however, was lower than that noted in the control plants (kept at 24°C for 72 h). Since the detailed conditions for the plant growth and the cold treatment used by Maia et al. (1998) were not described, the cause of this discrepancy is unclear

1165

1166


Kozak, L.P., Britton, J.H., Kozak, U.C. and Wells, J.M. (1988) The mitochondrial uncoupling protein gene. Correlation of exon structure to transmembrane domains. / . Biol. Chem. 263: 12274-12277. Laloi, M., Klein, M., Riesmeier, J.W., Miiller-Rober, B., Fleury, C , Bouillaud, F. and Ricquier, D. (1997) A plant cold-induced uncoupling protein. Nature 389: 135-136. Liu, X., Bell, A.W., Freeman, K.B. and Shore, G.C. (1988) Topogenesis of mitochondrial inner membrane uncoupling protein. Rerouting transmembrane segments to the soluble matrix compartment. J. Cell. Biol. 107: 503-509. Lutcke, H.A., Chow, K.C., Mickel, F.S., Moss, K.A., Kern, H.F. and Scheele, G.A. (1987) Selection of AUG initiation codons differs in plants and animals. EMBO J. 6: 43-48. Maia, I.G., Benedetti, C.E., Leite, A., Turcinelli, S.R., Vercesi, A.E. and Arruda, P. (1998) AtPUMP: an Arabidopsis gene encoding a plant uncoupling mitochondrial protein. FEBS Lett. 429: 403-406. Murashige, T. and Skoog, K. (1962) A revised medium for plant growth and bioassays with tobacco culture. Physiol. Plant. 15: 473-497.

Nicholls, D.G. and Locke, R.M. (1984) Thermogenic mechanisms in brown fat. Physiol. Rev. 64: 1-64. Page, R.D.M. (1996) TREE-VIEW: an application to display phylogenetic trees on personal computers. Comput. Appl. Biosci. 12: 357-358. Ricquier, D., Casteilla, L. and Bouillaud, F. (1991) Molecular studies of the uncoupling protein. FASEB J. 5: 2237-2242. Seymour, R.S. (1997) Plants that worm themselves. Sci. Amer. 276: 104109. Solanes, G., Vidal-Puig, A., Grujic, D., Flier, J.S. and Lowell, B.B. (1997) The human uncoupling prtoein-3 gene. Genomic structure, chromosomal localization, and genetic basis for short and long form transcripts. J. Biol. Chem. 272: 25433-25436. Thompson, J.D., Higgins, D.G. and Gibson, T.J. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucl. Acid Res. 22: 4673-4680. Vercesi, A.E., Martins, I.S., Silva, M.A.P., Leite, H.M.F., Cuccovia, I.M. and Chaimovich, H. (1995) PUMPing plants. Nature 375: 24.

(Received April 28, 1999; Accepted September 6, 1999) Downloaded from http://pcp.oxfordjournals.org/ by guest on October 26, 2016