The Plant Cell, Vol. 4, 1539-1548, December 1992 O 1992 American Society of Plant Physiologists

Pumpkin Phloem Lectin Genes Are Specifically Expressed in companion CeIIs

Dwight E. Bostwick, 'Joanne M. Dannenhoffer,' Megan 1. Skaggs,' Richard M. ListerYb Brian A. Larkins,' and Gary A. Thompsonai'

a Department of Plant Sciences, University of Arizona, Tucson, Arizona 85721 Department of Botany and Plant Pathology, Purdue University, West Lafayette, Indiana 47907

Pumpkin phloem exudate contains two abundant phloem proteins: PP1 is a 96-kD protein that forms polymeric filaments in vivo, and PP2 is a 48-kD dimeric lectin. Polyclonal antibodies raised against pumpkin phloem exudate were used to isolate severa1 cDNAs correspondlng to PP1 and PP2. RNA gel blot analysis indicated that PP1 is encoded by an mRNA of -2500 nucleotides, whereas PP2 subunits are encoded by an mRNA of 1000 nucleotides. Sequence analysis of PP2 cDNAs revealed a 654-bp open reading frame encoding a 218-amino acid polypeptide; this polypeptide had the carbohy- drate binding characteristics of a PP2 subunit. The PP2 mRNA was localized within the phloem of pumpkin hypocotyl cross-sections based on in situ hybridization of a digoxigenin-labeled antisense probe. PP2 mRNA was found within the companion cells in both the bicollateral vascular bundles and the extrafascicular phloem network.

INTRODUCTION

Protein inclusions have been observed in phloem tissue of di- cotyledonous plants for many years using light microscopy (reviews by Cronshaw and Sabnis, 1990; Evert, 1990). The term P-protein (phloem protein) was introduced to describe these proteinaceous filaments and aggregations observed in sieve elements by transmission electron microscopy (Cronshaw and Esau, 1967; Esau and Cronshaw, 1967). These proteins are a major component of the cytoplasmic contents of sieve ele- ments; they are synthesized very early in phloem ontogeny and persist in senescent sieve elements. Although P-proteins are a major component of sieve elements, P-protein bodies have also been observed in companion cells and phloem parenchyma cells of developing cucurbit phloem tissue (Cron- shaw and Esau, 1968a, 1968b).

Phloem exudates from Cucurbita species contain high con- centrations of P-protein filaments and have been a useful source for the biochemical analysis of P-proteins. Phloem ex- udates from pumpkin contain two abundant proteins with similar properties (Kollmann et al., 1970; Beyenbach et al., 1974; Read and Northcote, 1983). The two proteins, PP1 (phloem protein 1) and PP2 (phloem protein 2), are basic polypeptides that have similar amino acid compositions and are components of phloem filaments (Beyenbach et al., 1974; Weber et al., 1974). The PP1 monomer is a96-kD protein that cross-links with other PP1 monomers by disulfide linkages, forming soluble polymers in vivo (Walker, 1972; Beyenbach et al., 1974; Sabnis and Hart, 1979; Read and Northcote, 1983). Upon oxidation, PP1 poly- mers cross-link to form an insoluble matrix or gel (Walker, 1972;

To whom correspondence should be addressed.

Read and Northcote, 1983). PP1 appears to be the primary structural protein involved in the formation of slime plugs that are seen at sieve plates in electron micrographs of disrupted vascular tissues (Walker and Thaine, 1971). PP2 is a 48-kD dimeric lectin that preferentially binds oligomers of N-acetyl- D-glucosamine (Beyenbach et al., 1974; Sabnis and Hart, 1978; Allen, 1979; Read and Northcote, 1983). Unlike PP1, purified PP2 is soluble when exposed to either atmospheric oxygen or oxidizing agents. Within the phloem filaments, PP2 is cova- lently linked to PP1 by disulfide bridges (Kleinig et al., 1975; Read and Northcote, 1983).

Many unresolved questions exist regarding the function of P-proteins and the physiological relationships between phloem cell types. An intriguing relationship exists between the sieve elements and their respective companion cells. The mature sieve element-companion cell complex contains two onto- genetically related cell types that are morphologically and physiologically distinct, yet appear to be functionally interac- tive. Within the two cell types, P-proteins have been observed in developing companion cells and sieve elements as well as mature sieve elements. Because the nucleus and ribo- somes degenerate during the maturation of the sieve element, P-protein synthesis could occur either in the immature sieve elements, companion cells, or both. Resolution of these ques- tions requires more understanding of.the cell specificity and developmental timing of P-protein synthesis.

We are interested in understanding the developmental regu- lation of P-protein gene expression in pumpkin to ultimately address questions about P-protein synthesis, accumulation, interactions, and function in phloem tissue. We raised

1540 The Plant Cell

polyclonal antibodies against the total phloem exudate frompumpkin seedlings and used them to isolate cDNAs corre-sponding to P-proteins. One group of cDNAs contained an openreading frame (ORF) that encoded a polypeptide with the phys-ical and functional characteristics of PP2. Several partial cDNAclones were also isolated that have features correspondingto PP1 mRNA. In situ hybridization of antisense probes local-ized the PP2 mRNA within the companion cells of phloemtissue in pumpkin hypocotyls.

RESULTS

Isolation of P-Protein cDNAs

To obtain cDNA clones corresponding to PP1, PP2, and addi-tional pumpkin P-proteins, we raised polyclonal antibodies inchickens against total reduced proteins from pumpkin phloemexudate. A complex antiserum was obtained that reacts withmany of the phloem exudate proteins resolved by SDS-PAGE,as shown in Figure 1 (lane 2 of SDS-PAGE and lane 1 of theimmunoblot). To determine if the antiserum was specific forphloem exudate proteins, we tested it for cross-reactivity withproteins isolated from pumpkin callus tissue (Figure 1, lane3 of SDS-PAGE and lane 2 of the immunoblot). Although

IgY

O UJ O UJ O

2114KD —

107.0 kD—

«8.3kD —

45.«kD —

11.2 U> —

15.4 kO —

••• ^^

SOS-PAGE Immunoblot

Figure 1. SDS-PAGE and Immunoblot Analysis of Pumpkin PhloemExudate and Callus Proteins.Pumpkin phloem exudate protein (lane 2, SDS-PAGE; lanes 1 and 3,immunoblot) or total protein isolated from pumpkin callus tissue (lane3, SDS-PAGE; lanes 2 and 4, immunoblot) were electrophoresed ina 12.5% polyacrylamide gel. The proteins (30 ng per lane) separatedby SDS-PAGE were stained with Coomassie blue. For immunoblot anal-ysis, the proteins (3 ng per lane) were transferred from the gel ontonitrocellulose and incubated with either the total antiserum raisedagainst phloem proteins or preimmune serum from the same hen.

SDS-PAGEof PhloemExudate

PP1-

PP2-

5' 8' 8' 5'Developing Times

Figure 2. Immunoblot of Exudate Proteins with Purified Antiserum.

The first lane is a Coomassie blue-stained SDS-polyacrylamidegel of total phloem proteins (30 ng). Three micrograms of phloemexudate protein was loaded in each of the remaining lanes. After elec-trophoresis, these proteins were transferred onto a nitrocellulosemembrane; the membrane was cut into strips, and individual stripswere incubated with antiserum that was affinity purified with fusionproteins encoded by cPC7 (lane 7), cPC13 (lane 13), and CPC20 (lane20). The last lane (Total IgY) was incubated with the antiserum raisedagainst total phloem exudate proteins. The arrows indicate immunoreac-tive bands, and the time allowed for color development for eachimmunoreaction is indicated below each lane in minutes.

callus tissue contains a large number of abundant proteins,the antiserum cross-reacted with only a single protein band.This protein has the mobility of PP1 and may reflect PP1synthesis in differentiating phloem cells within the callus,because P-proteins have previously been observed in differen-tiating sieve elements in squash callus tissue (Lackney, 1991).The absence of cross-reactivity between the antiserum andproteins from callus tissue reflects its specificity for the phloemexudate proteins. The preimmune serum did not cross-reactwith proteins from pumpkin phloem exudate or pumpkin callus.

To identify mRNAs corresponding to phloem proteins, anexpression cDNA library was constructed with poly(A)+ RNAisolated from pumpkin seedlings. This library was screenedwith the phloem protein antiserum and 22 immunopositiveplaques were obtained; the 10 most immunoreactive phageswere selected for further analysis. To determine if these 10clones represented unique or related sequences, we excisedthe pBluescript SK- plasmid containing each cDNA from

In Situ Localization of PP2 mRNA 1541

the X ZAP clone, purified the DNA, and performed a cross-hybridization analysis. The results of these experiments showedthat the 10 cDNAs represented two groups of closely relatedor identical sequences (data not shown). Clones with the largestcDNA inserts were selected for detailed analysis and weredesignated cPC7 (1.38 kb) and CPC13/20 (cPC13 is 980 bp andcPC20 is 792 bp).

Identification of the Proteins Encoded byPhloem cDNAs

To determine the type of P-protein encoded by each clone, thep-galactosidase fusion proteins were used to affinity purify spe-cific antibodies from the complex antiserum. Each of thepurified antibodies was reacted with a protein blot of totalphloem exudate. Figure 2 shows the result of one such analy-sis; cPC7 encodes a polypeptide that is immunologicallyrelated to PP1 (lane 7), whereas both cPC13 and cPC20 en-code a polypeptide that is immunologically related to PP2(lanes 13 and 20).

As seen in Figure 3, RNA gel blot hybridization of total andpoly(A)+ RNA from pumpkin seedlings with cPC7 and cPC13

Total Poly A*10 ug 50 ng

Total Poly A*10 ug 50 ng

1.77kb —1.52kb —1.28 kb —

0.78 kb —

0.53 kb —

0.40 kb —

0.28 kb —

0.16kb —

CPC20 cPC7

Figure 3. RNA Gel Blot Analysis of Total and Poly(A)+ RNA fromPumpkin Seedlings.Total RNA (10 ng per lane) and poly(A)* RNA (50 ng per lane) wereisolated from pumpkin seedlings and separated by electrophoresisin an agarose-glyoxal gel. The RNA was transferred onto a Nytran mem-brane, divided into two sections, and hybridized with 32P-labeledinserts from cPC20 and cPC7.

MDNKEKEARE KLGGEVKLGH £LDVILKNAD

VALHYPSFLK LYDQLVAGIL LNKGAIKYIF

DKKSNSNWYF IFARALSIAW IEDKRYWKWG

S£GDSNVAEL IEVSWLDIRG KINESMLSQN

WYEVALQVQ LNSRASGWNA PLNIELKKPD

GSKIARQEQL LGKPQNQWFE IWEFKVGNH

GC.GSSGEIEF AFFEHGGHWK RGLLVKGVRI

GAKG£G£A

Figure 4. Deduced Amino Acid Sequence of the PP2 Subunit.The amino acid sequence deduced from CPC13/20 is 218 amino acidsin length and appears to encode an entire PP2 subunit. The positionsof the six cysteine residues, involved in disulfide linkages between thetwo subunits of the PP2 dimer and linking PP2 to PP1 filaments, areunderlined.

showed that the corresponding mRNAs are abundant and dis-tinct species. cPC13 hybridized to a single mRNA of ~1000nucleotides, which agrees with previous measurements of themRNA encoding PP2 (Sham and Northcote, 1987). In contrast,cPC7 hybridized to an mRNA of ~2500 nucleotides corre-sponding to the calculated size of an mRNA encoding PP1.This clone encodes ~1380 nucleotides and is not a full-lengthcopy of the PP1 mRNA.

The results of the immunoblot and RNA gel blot analysessuggested that CPC13/20 encoded PP2. Nucleotide sequenceanalysis of CPC13/20 revealed an ORFof 654 nucleotides. TheORF encodes an entire PP2 subunit; the deduced polypep-tide of 218 amino acids shown in Figure 4 has a calculatedmolecular mass of 24,478 D. As shown in Table 1, this poly-peptide is rich in glycine (10.1%), leucine (10.1%), and lysine(9.6%). The amidated and acidic amino acids account for 20.6%(Asx 10.1%, Glx 10.5%) of the total amino acid content. As-paragine is more prevalent than glutamine (Asn 6.0%, Gin3.2%), whereas glutamic acid is more abundant than aspar-tic acid (Glu 7.3%, Asp 4.1%). The polypeptide has six cysteineand no threonine residues. The six cysteine codons ofCPC13/20 agree with the six cysteine residues identified withinfully reduced PP2 polypeptides by biochemical methods(Beyenbach et al., 1974; Read and Northcote, 1983). The ab-sence of threonine codons concurs with the amino acidanalysis of the squash phloem lectin reported by Alien (1979),but contrasts with the analysis reported by Beyenbach et al.(1974) who corrected the serine and threonine values for decom-position. We also compared the amino acid composition ofthe deduced polypeptide with experimentally determinedamino acid compositions of phloem lectins from pumpkin(Beyenbach et al., 1974) and squash (Alien, 1979). All threehave similar amino acid compositions (Table 1).

To demonstrate carbohydrate binding activity of the poly-peptide encoded by cPC13/20, we inserted the entire ORF,including the translation termination codon, into a T7

1542 The Plant Cell

Table 1. Deduced and Observed PP2 Amino Acid Compositions of cPC13/20 and Previously Published Biochemical Quantitations

Deduced No. aa Allen (1979)b Beyenbach et al. (1974)b cPC13/20 Allen (1979) Beyenbach et al. (1974) Amino Acid in cPC13/20a (mollmol) (mollmol) (mo1 O/o) (mo1 O/o) (mo1 O/O)

Ala 16 14.4 16.4 7.3 6.6 7.5 Arg 8 8.4 7.6 3.7 3.8 3.5 ASP 9 21 .6c 24.2c 4.1 9.9 11.1 Asn 13 6.0 CYS 6 6.0 4.4 2.8 2.7 2.0 Glu 16 23.OC 22.2c 7.3 10.4 10.2 Gln 7 3.2 GlY 22 24.0 22.2 10.1 11.0 10.2 His 5 4.8 5.2 2.3 2.1 2.4 lle 15 14.3 11.6 6.9 6.6 5.3 Leu 22 22.8 22.0 10.1 10.4 10.1 LYS 21 19.2 21 .I 9.6 8.8 9.7 Met 2 2.4 2.2 0.9 1.1 1 .o Phe 9 8.4 8.7 4.1 3.8 4.0 Pro 4 4.8 7.8 1.8 2.2 3.6 Ser 14 14.4 12.2 6.4 6.6 5.6 Thr O 0.0 6.3 0.0 0.0 2.9 TrP 8 4.8 5.7 3.7 2.2 2.6 TY r 6 7.2 5.9 2.8 3.3 2.7 Val 15 18.0 11.3 6.9 8.2 5.2

a aa, amino acids. b Published values were adjusted to reflect a deduced polypeptide of 218 amino acids. c Allen and Bevenbach et al. reported values for Asx and Glx.

expression vector, pRSETB, and synthesized a fusion protein in Escherichia coli. Synthesis of the fusion protein was initiated from the pRSETB vector sequence, which increased the mo- lecular size of the fusion protein by 3740 D. Following induction, the fusion protein accumulated to high levels in the bacteria, as shown in Figure 5A. The fusion protein was affinity purified to homogeneity from the crude bacterial lysate by batch ad- sorption to ovomucoid-acryl beads. Elution from the affinity matrix with 1 mM tri(N-acetyl-o-glucosamine) resulted in a quantitative recovery of the protein (Figure 58). The pRSET vector also encodes a metal binding protein domain that is expressed as the N-terminal portion of the recombinant protein. A fusion protein encoded by the chloramphenicol acetyltransferase (CAT) gene in pRSET also contains the metal binding domain. However, the CATfusion protein did not bind to the ovomucoid affinity matrix, demonstrating that binding was specific to the polypeptide region of the fusion protein en- coded by cPC13/20 and not the metal binding domain.

In Situ Hybridization of PP2 mRNA

Previous studies have shown that the phloem lectin occurs in both sieve elements and companion cells (Smith et al., 1987). To obtain further evidence that the cDNA clone we isolated corresponded to a phloem-specific protein and to determine its site of synthesis, we localized PP2 mRNA by in situ

hybridization. Cross-sections of pumpkin hypocotyl tissue were incubated with in vitro-synthesized transcripts labeled with digoxigenin-11-UT!? The use of this nonisotopic labeling method allowed us to achieve high spatial resolution of the signal with retention of tissue morphology. Figure 6 shows the complex phloem anatomy of the pumpkin seedling hypocotyl. The Cucurbitaceae is one of severa1 plant families that have bicollateral vascular bundles composed of interna1 and exter- na1 phloem (fascicular phloem). A second feature that adds to the complexity of cucurbit phloem anatomy is the extrafas- cicular phloem, which occurs in strands within the cortex and in arcs bordering both sides of the bundle (Crafts, 1932; Blyth, 1958). In addition to the primary phloem, secondary phloem within the vascular bundle is derived from a vascular cambium.

PP2 antisense transcripts hybridized to mRNA within the phloem of hypocotyl tissues and are visible as a blue-purple precipitant after the alkaline phosphatase reaction, as seen in Figure 7. PP2 transcripts occurred in both the bundle and extrafascicular phloem tissue, although labeling of individual sections was variable. In certain sections, label was only de- tected in the extrafascicular arc of the phloem (Figure 7A), whereas, in others, signal appeared in both the bundle phloem and the extrafascicular phloem (Figure 78). In addition, the extrafascicular phloem strands within the cortex were often labeled (Figures 7C and 7D).

The nonisotopic labeling method can detect transcripts within individual cell types using light microscopy. We identified sieve

In Situ Localization of PP2 mRNA 1543

MHours after induction

0 1 2 3 4

BM

Figure 5. Carbohydrate Binding Activity of the Fusion Protein Encodedby cPC13.

The ORF encoded by cPC13 was amplified by PCR and inserted intothe protein expression vector pRSETB (Invitrogen).(A) A Coomassie-stained SDS-polyacrylamide gel of the £ coli crudelysate at 0, 1, 2, 3, 4, and 5 hr following the induction of the fusionprotein. The letter M designates molecular size markers. In descend-ing order, the molecular size markers are 106, 80,49.5, 32.5,27.5, and18.5 kD. Synthesis of the fusion protein increases the molecular sizeof PP2 by 3.74 kD.(B) A silver-stained SDS-polyacrylamide gel of PP2 and PP2 fusionprotein affinity purification by ovomucoid-acryl beads. In separate reac-tions, total crude phloem exudate from pumpkin seedlings (lane 1) andcrude £. coli lysate containing the PP2 fusion protein (lane 3) wereadsorbed onto ovomucoid-acryl beads. Purified PP2 (lane 2) and fusionPP2 (lane 4) were eluted from the matrix with 1 mM tri(W-acetyl-D-glucosamine). Crude E. coli lysate containing the CAT fusion protein(lane 5) did not bind to the affinity matrix (lane 6). The letter M indi-cates molecular size markers. In descending order, the molecular sizemarkers are 106, 80, 49.5, 32.5, 27.5, and 18.5 kD.

elements, companion cells, and phloem parenchyma withinthe phloem of hypocotyls 12 days after germination. PP2 mRNAwas only localized in companion cells within both bundles andextrafascicular phloem (Figures 7C to 7E).

PP2 sense transcripts were used as a control for nonspecificbinding of probe to the tissue. Control sections showed no sig-nal and the protoplasts stained greenish brown (Figure 7F).

To detect false positives (i.e., alkaline phosphatase activity notdue to hybridization of the probe), we included two additionalcontrols. First, to detect endogenous alkaline phosphatase ac-tivity, we did not add antibody after hybridization with senseor antisense transcripts. Second, sections without transcripts(i.e., no digoxigenin) were used to detect activity due to non-specific binding of the antibody to the tissues. No false positiveswere detected.

DISCUSSION

Although several hypotheses have been proposed regardingthe role of P-proteins in phloem cells, the function of these pro-teins is unknown. Early proposals that P-protein filaments playan active physical role in translocation, or that P-proteins maybe degradation products formed during the maturation of sieveelements, have generally been discarded (Cronshaw, 1975).Current hypotheses suggest that the lectin (PP2) binds toglycoconjugates of either the sieve element reticulum or theplasma membrane, thereby anchoring the P-protein filaments(Sabnis and Hart, 1978; Smith et al., 1987). The loss of hy-drostatic pressure that occurs upon phloem wounding disruptsthe lectin-carbohydrate interaction releasing the filaments fromtheir parietal positions into the assimilate stream. The filamentscould block the flow of assimilates at the sieve plate by form-ing slime plugs and oxidize at the wound surface to seal thewound (Sabnis and Hart, 1979). A variation of this hypothesissuggests that PP1 filaments form a gel creating a physical

v

Figure 6. Toluidine Blue-Stained Cross-Section of Pumpkin Hypocotyl.

C, collenchyma; EP, external fascicular phloem; IP, internal fascicularphloem; X, xylem; arrowheads, extrafascicular phloem, both corticalstrands and arcs bordering the bundle phloem. Bar = 200 urn.

1544 The Plant Cell

v

EP

4-JfX IP

>

EP

4*C

.,<•£ V^ *^

IP

IP

Figure 7. In Situ Localization of PP2 mRNA in Pumpkin Hypocotyl Cross-Sections.

SE

flSExPP

E SP

In Situ Localization of PP2 mRNA 1545

barrier and that the associated PP2 lectins bind fungi and bac- teria as a biochemical barrier to microbial ingress into the wounded tissue (Read and Northcote, 1983). There is little experimental evidence to either support or reject these hypoth- eses, because assessing the function of these proteins is technically complex and the tools to perform these analyses have not been available. The molecular probes we have de- veloped will allow us to investigate the roles of these proteins in the phloem.

The analysis of the protein encoded by cPC13/20 is consis- tent with the conclusion that these clones encode the phloem lectin. Assuming this is true, one would predict that the corre- sponding mRNA transcripts would originate from either sieve elements, companion cells, or both. The use of nonisotopic labeling methods for in situ localization of mRNA allowed us to localize PP2 mRNA, with a high degree of resolution, within the companion cells in seedling hypocotyls. The localization of this mRNA parallels the pattern of PP2 distribution in bun- dle and extrafascicular phloem previously reported by Smith et al. (1987). They found that polyclonal lectin antibodies bound predominantly to the P-protein of the cortical extrafascicular phloem and in the arcs of extrafascicular phloem located on either side of vascular bundles. Our localization of PP2 mRNA in extrafascicular phloem clearly mimics the localization of the protein in both the extrafascicular arcs and the cortical strands. In addition, Smith et al. (1987) reported that the fascicular phloem of mature stems stained lightly, while the fascicular phloem of older stems stained strongly. We also detected tran- scripts in the fascicular phloem, although less often than in the extrafascicular phloem.

We detected PP2 mRNA only in the companion cells, whereas PP2 protein was immunocytochemically localized in both the companion cells and sieve elements (Smith et al., 1987). 60th of these results are consistem with our current knowledge of sieve element and companion cell contents. Gietl and Ziegler (1979) found no evidence of mRNA or other com- ponents of the translational machinery in exudates from mature sieve elements. Because the nucleus and ribosomes degener- ate during sieve element maturation, it is not surprising that we did not detect PP2 mRNA in these cells.

Based on the abundance of P-protein in the mature sieve elements, it is easy to envision transcription and protein synthesis within the immature sieve elements. P-protein bod- ies have been observed in immature sieve elements and companion cells of cucurbits (Cronshaw and Esau, 1968a).

The perplexing issue is the accumulation of protein and mRNA within mature companion cells. The presence of PP2 mRNA and protein in mature companion cells suggests that protein synthesis may be active within a cell type where P-proteins are not expected to accumulate. Smith et al. (1987) speculated that PP2 could be synthesized in the companion cells and subsequently transported into the sieve elements via plas- modesmata. Our results support this hypothesis. However, the localization of PP2 mRNA in seedling hypocotyl tissue was done at a single stage of development where the primary phloem tissue was mature. Thus, we cannot eliminate the pos- sibility that PP2 synthesis occurs in both cell types during early stages of cell differentiation. We are currently using in situ hy- bridization and immunocytochemical techniques to examine the temporal and spatial pattern of PP2 gene expression dur- ing phloem differentiation.

Biochemical analyses of PP2 have shown that this protein is a dimer and has carbohydrate binding characteristics typi- cal of plant lectins. Read and Northcote (1983) showed that PP2 is composed of two subunits, a (M, 26,500) and p (Mr 25,000), joined by disulfide linkages between cysteine residues. However, separation of the subunits could oniy be observed when iodoacetamide-treated PP2 dimers were fully reduced and separated on SDS-polyacrylamide gels at pH 9.2. Com- paring the deduced and experimentally determined amino acid content of PP2 suggests that the two subunits are similar. Whether the two subunits are encoded by different genes or i f their differences in mobility by SDS-PAGE are due to tran- scriptional processing or post-translational modification of a single gene product is unknown. In addition, little is known about the specific protein-carbohydrate interactions that oc- cur between PP2 and the chitin ligand. A conserved 43-amino acid domain that is typical of most chitin binding proteins (Chrispeels and Raikhel, 1991) could not be identified in PP2; thus, PP2 appears to be structurally distinct from other chitin binding lectins.

METHODS

Materials

Restriction endonucleases were from Bethesda Research Laborato- ries. u-~*-P-~ATP or dCTP (3000 Cilmmol) and u -~~S-~ATP (1000 to 1500 Cilmmol) were from DuPont-New England Nuclear. Oligonucle-

Figure 7. (continued).

C, collenchyma; CC, companion cell; EP, external fascicular phloem; IP, interna1 fascicular phloem; PP, phloem parenchyma; SE, sieve element; SP, sieve plate; X, xylem; arrowheads, extrafascicular phloem. (A) to (E) Sections hybridized to cPC20 antisense strand transcripts. (F) Section hybridized to cPC20 sense strand transcripts. (A), (B), and (F) Cross-sections through vascular bundle and surrounding areas. Bars = 50 pm. (C) Section of the outer cortex (collenchyma indicated for reference) with extrafascicular phloem strands. Bar = 25 pm. (D) Section through extrafascicular strand showing sieve element and companion cell in longitudinal section. Bar = 25 pm. (E) Portion of external bundle phloem. Bar = 10 pm.

1546 The Plant Cell

otides were synthesized by the University of Arizona Macromolecular Structure Facility, Tucson. Random primer radioactive labeling kits were from Ambion (Austin, TX). Polymerase chain reaction (PCR) reagent kits were from Perkin Elmer Cetus. Nitrocellulose and Nytran mem- branes were from Schleicher & Schuell.

lsolation of Pumpkln Phloem Exudate Proteins and Callus Proteins

Pumpkin (Cucurbifa maxima Duch. cv Big Max) seedlings were grown at 25OC with a 16-hr photoperiod. At 10 to 12 days after germination, seedling hypocotyls were cut and microliter quantities of phloem exu- date were collected in micropipettes. The exudate was diluted 1:4 in a deaerated extraction buffer consisting of 0.1 M Tris, pH 8.2, 5 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride, and either 100 mM 2-mercaptoethanol or 20 mM DTT (Read and Northcote, 1983). Pumpkin callus cultures were initiated from surface-sterilized leaf tissue on a medium consisting of MS salts (Murashige and Skoog, 1962), 3% su- crose, 0.4 mg L-l thiamine, 100 mg L-l myo-inositol, 0.5 pM kinetin, 13.6 mM 2,4-D, pH 5.6, and 0.6% agar. After 4 weeks, the callus was transferred to a similar medium containing 5 pM benzyladenine and 0.5 pM 2,4-D (Kim et al., 1988). These cultures were grown under con- stant illumination at 22OC. Pumpkin callus tissue (0.5 g) was homogenized in 500 pL of extraction buffer containing 100 mM 2-mercaptoethanol. The cellular debris was pelleted by centrifugation at 10.000 rpm for 5 min, and the supernatant was collected.

lsolation of Phloem Exudate Antlbodies

Antibodies were raised in laying hens against the total phloem exu- date isolated from pumpkin seedlings. A 100-pL aliquot of the diluted phloem exudate sample (-750 pg of total protein) was emulsified with an equal volume of Freunds complete adjuvant and injected subcutane- ously. Two subsequent injections with an equal amount of phloem exudate and Freunds incompiete adjuvant were made at 2-week in- tervals. Eggs were collected 1 week following the final injection, and antibodies were purified from egg yolks as described by Carroll and Stoller (1983) and Song et al. (1985). Antibody preparations were used at dilutions of 1:2500.

Electrophoresls and Blotting of Proteins

Total reduced phloem exudate proteins and reduced soluble callus proteins were separated by SDS-PAGE in a 12.5% acrylamide gel (Laemmli, 1970). Following electrophoresis, the gel was cut and si- ther stained with Coomassie Brilliant Blue R 250 or the proteins were transferred from the gel to nitrocellulose filters by electroblotting in a TransBlot apparatus (Bio-Rad) using a Tris-glycine buffer (lowbin

-et al., 1979). lmmunoblots were incubated with the complex phloem exudate antiserum and visualized by alkaline phosphatase-con- jugated rabbit anti-chicken (RAC) IgG (Jackson Immunoresearch Laboratories, Inc., West Grove, PA). Color was developed using 5-brome 4-chloro-3-indolyl phosphate p-toluidine salt (BCIP) and p-nitro blue tetrazolium (N BT).

Constructlon and Screenlng of a cDNA Expresslon Llbrary

Total RNA was isolated from pumpkin seedlings 7 days after germina- tion by the guanidinium thiocyanate method of MacDonald et al. (1987).

Poly(A)+ RNA, purified on Hybond mAP paper (Amersham Interna- tional), was used as the template for cDNA synthesis (PharmaciacDNA synthesis kit). EcoRllNotl adapters were ligated to the double-stranded cDNA, and the cDNAs were inserted into the EcoRCdigested bacte- riophage A. ZAP II (Stratagene). The bacteriophage were packaged with Gigapack II Gold packaging extract (Stratagene) and grown in XL1-Blue Escherichia coli host cells. The initial cDNA library contained -5.6 x 105 recombinant plaques and was amplified to an approxi- mate titer of 2.3 x 107 plaque-forming units per pL. A total of 5 x 104 phage clones were grown in XL1-Blue host cells, and P-galactosidase fusion proteins were induced by placing nitrocellulose filters im- pregnated with 50 mM isopropyl P-o-thiogalactoside onto the plated bacteriophage. The filters were incubated with the phloem exudate antiserum, and plaques immunoreactive with RAC antibodies con- jugated to alkaline phosphatase were visualized by BCIP/NBT color development. lmmunopositive plaques were purified, and the pBlue- script SK- phagemids containing the cDNAs were excised according to the manufacturer's instructions.

Affinity Puriflcation of P-Proteln Antlbodies

Plaque-purified h ZAP II bacteriophage containing the individual cDNAs cPC7, cPC13, or cPC20 were each grown to confluence on separate plates. The expression of fusion proteins was induced by placing a nitrocellulose filter impregnated with 50 mM isopropyl P-D-thiogalac- toside on each plate at 37%. The nitrocellulose filters with the bound fusion proteins were washed with TTBS (10 mM Tris, pH 7.4, 140 mM NaCI, 0.15% [v/v] Tween 20), blocked with 1% gelatin in TTBS, and incubated with the complex antiserum (1:2500) diluted in TTBS. After the nonspecific antibodies were removed with severa1 washes of 1% Triton X-100 in TTBS, the specific antibodies were eluted from the filters with 0.2 M glycine, pH 2.8, 1 mM EGTA, neutralized with 0.1 volume of 1 M Tris base, and concentrated by precipitation with an equal vol- ume of 4 M (NH4)2S04. The pellet was resuspended in 1 mL of 10 mM NaH2P04, pH 7.5, and dialyzed overnight against 10 mM NaH2P04, pH 7.5, and 0.01% NaN3 at 4OC. lmmunoblot analysis and visualiza- tion were conducted as described above.

FINA lsolation and Blottlng

Total RNA was extracted from hypocotyl tissue of pumpkin seedlings 10 to 12 days after germination using a scaled-down guanidine-HCI isolation method originally described by Cathala et al. (1983). Poly(A)+ RNA was purified with the PolyATract mRNA lsolation System 111 (Promega). RNA samples were denatured with glyoxal (Covey and Hull, 1981) and separated in 1.3% agarose gels with 1 x TAE (40 mM Tris- acetate, 1 mM EDTA) as the gel and electrode buffer. The RNA was transferred from the gel with 10 mM NaOH onto a Nytran membrane (Sambrook et al., 1989). The membrane, containing the RNA, was UV cross-linked (1.2 x 105 wJ/cm2) and incubated at 80°C for 1 hr. RNA blots were incubated for 30 min at 68OC in 10 mL of hybridization buffer (0.25 M NaP04, pH 7.2, 0.25 M NaCI, 1 mM EDTA, and 7% SDS). Denatured DNA probes (a-32P-dATP or dCTP-labeled PCR-amplified cPC13 open reading frame [ORF] or the EcoRl insert of cPC7) were added to the hybridization buffer, and the filters were incubated for 12 to 15 hr at 68OC. After hybridization, the filters were washed three times (5.60, and 60 min) at 68OC in a 1 x SSC (0.15 M NaCI, 0.015 M sodium citrate) and 0.1% SDS solution, and hybridization was visual- ized by autoradiography.

In Situ Localization of PP2 mRNA 1547

DNA Sequence Analysis

Enzymatic sequencing was accomplished by the dideoxy chain ter- mination method according to the instructions accompanying the Sequenase version 2.0 sequencing kits (U.S. Biochemicals). Com- puter-aided analysis of the nucleotide and amino acid sequences was performed with programs supplied by the Genetics Computer Group (Madison, Wl), Microgenie computer program (Beckman Instruments, Inc., Palo Alto, CA), and Genejockey computer program (Biosoft, Cam- bridge, UK).

Synthesls ot PP2 Fuslon Protelns

A 678-bp sequence of cPC13 that included the complete PP2 ORF was amplified by PCR. Two 25-bp oligonucleotide primers were de- signed from the sequence of cPC13: the 5' oligonucleotide contained the translation initiating methionine codon and the 3'oligonucleotide contained the translation stop codon. The ORF was amplified for 30 cycles (95OC for 1 min, 55OC for 1 min, 72OC for 1 min) and ligated in frame into the protein expression vector pRSETB (Invitrogen, San Diego, CA). The construction was sequenced to confirm the main- tenance of the reading frame, and fusion protein was synthesized according to the manufacturer's instructions.

Lectln Blnding Assay

The lectin binding assay was based on the method described by Read and Northcote (1983). Five hundred milligrams of ovomucoid-acryl beads (Sigma) was equilibrated in binding buffer (50 mM Tris, pH 8.2, 500 mM NaCI, 3 mM NaN3, 2 mM EDTA, 0.5% Triton X-100, 20 mM DTT). Twenty micrograms of total phloem protein or 25 pL of the fu- sion protein lysate in 100 pL of binding buffer was adsorbed to 20 pL of ovomucoid-acryl beads with constant mixing at 4OC for 2 hr. The supernatant was removed, and the affinity matrix was washed three times with 1 mL of the binding buffer. The purified PP2 was eluted from the matrix in 100 pL of 1 mM N',N",N"'-triacetylchitotriose (Sigma) with constant mixing at 4OC for 30 min. The eluate was removed and the matrix was washed two additional times with 100 pL of 1 mM tri- acetylchitdriose. The eluate and washes were combined, vacuum dried, and dissolved in 20 pL of distilled water.

In Sltu Hybrldization

Hypocotyl pieces of pumpkin seedlings 10 to 12 days after germina- tion were fixed at room temperature in 2% glutaraldehydel50 mM KPO.,, pH 7.0, for 3 hr, then dehydrated and embedded in paraffin. Paraffin blocks were sectioned at 10 pm and mounted on poly-L-lysine- coated slides. Digoxigenin-labeled sense and antisense riboprobes were synthesized by in vitro transcription from pBluescript KS+ (Strata- gene) that contained the cDNA template cPC20 (PP2). Plasmids were linearized and digoxigenin-11-UTP was incorporated with either T3 or T7 polymerase according to the manufacturer's instructions (Boehringer Mannheim). In general, the in situ hybridization technique followed the methods of Cox and Goldberg (1988). Transcripts were sheared to -150 bp by alkaline hydrolysis. Severa1 prehybridization treatments were incorporated to eliminate nonspecific binding of the probes and increase access of the probe to the target mRNA. The hypocotyl sec- tions were dewaxed, hydrated, and blocked in 1% BSA for 10 min. The slides were then sequentially treated with HCI, proteinase K, and acetic anhydride according to Bochenek and Hirsch (1990). The Sections

were prehybridized in 50% formamide, 4 x SSC, 1 x Denhardt's solution (1 x Denhardt's solution is 0.02% Ficoll, 0.02% PVP, 0.02% BSA), 0.5 mglmL salmon sperm DNA, 0.25 mglmL E. coli tRNA, 5% dextran sulfate, and 10 mM DTT for 4 hr at room temperature. Digoxigenin-labeled probe was added to an identical solution and ap- plied to slides for 14 to 16 hr at 42OC. Following hybridization, the sections were briefly washed in 4 x SSC and 5 mM DTT, incubated with 50 pglmL RNase A in NTE (500 mM NaCI, 10 mM Tris, pH 8.0, 1 mM EDTA) for 30 min at V C , and washed twice with NTE for 20 min. The slides were then washed four times in descending concen- trations of SSC (2 x, 1 x, 0.5 x, 0.1 x) with 1 mM DDT for 30 min each. The first three washes were at V C , and the final wash was at 42OC. Hybridization of the riboprobes was detected with anti- digoxigenin antibodies conjugated to alkaline phosphatase and visualized by color development with BClP and NBT (Boehringer Mannheim).

ACKNOWLEDGMENTS

The authors would like to thank the W. Atlee Burpee Company (War- minster, PA) for kindly donating the seed for these studies.

Received September 25, 1992; accepted October 21, 1992.

REFERENCES

Allen, A.K. (1979). A lectin from the exudate of the fruit of the vegeta- ble marrow (Cucurbita pepo) that has a specificity for a B-1,Clinked Kacetylglucosamine oligosaccharide. Biochem. J. 183, 133-137.

Beyenbach, J., Weber, C., and Klelnlg, H. (1974). Sieve-tube proteins from Cucurbita maxima. Planta 119, 113-124.

Blyth, A. (1958). Origin of primary extraxylary stem fibers in dicotyledons. Univ. Cal. Berkeley Publ. Bot. 30, 145-232.

Bochenek, B., and Hlrsch, A.M. (1990). In-situ hybridization of nodu- lin mRNAs in root nodules using non-radioactive probes. Plant MOI. Biol. Rep. 8, 237-248.

Carroll, S.B., and Stoller, B.D. (1983). Antibodies to calf thymus RNA polymerase I1 from egg yolks of immunized hens. J. Biol. Chem.

Cathala, G., Savouret, J.F., Mendez, B., West, B.L., Karln, M., Martial, J.A., and Baxter, J.D. (1983). Laboratory methods: A method for isolation of intact translationally active ribonucleic acid. DNA 2,

Chrlspeels, M.J., and Ralkhel, N.V. (1991). Lectins, lectin genes, and their role in plant defense. Plant Cell 3, 1-9.

Covey, S.N., and Hull, R. (1981). Transcription of cauliflower mosaic virus DNA. Detection of transcripts, properties, and location of the gene encoding the virus inclusion body protein. Virology 111,

Cox, K.H., and Goldberg, R.B. (1988). Analysis of plant gene expres- sion. In Plant Molecular Biology: A Practical Approach. C.H. Shaw, ed (Oxford: IRL Press), pp. 1-35.

Cratts, A.S. (1932). Phloem anatomy, exudation and transport of or- ganic nutrients in Cucurbits. Plant Physiol. 7, 183-225.

258, 24-26.

329-335.

463-474.

1548 The Plant Cell

Cronshaw, J. (1975). P-proteins. In Phloem Transport. S. Aronoff, J. Dainty, P.R. Gorham, L.M. Srivastave, and C.A. Swanson, eds (New York: Plenum Press), pp. 79-115.

Cmnshaw, J., and Esau, K. (1967). Tubular and fibrillar components of mature and differentiating sieve elements. J. Cell Biol. 34,801-815.

Cronshaw, J., and Esau, K. (1968a). P-Protein in the phloem of Cu- CurOita. I. Thedevelopment of P-protein bodies. J. Cell Biol. 38,25-39.

Cronshaw, J., and Esau, K. (1968b). P-Protein in the phloem of Cu- curbita. 11. The P-protein of mature sieve elements. J. Cell Biol. 38,

Cmnshaw, J., and Sabnls, D.D. (1990). Phloem proteins. In Sieve Ele- ments: Comparative Structure, lnduction and Development. H.-D. Behnke and R.D. Sjolund, eds (New York: Springer-Verlag), pp.

Esau, K., and Cronshaw, J. (1967). Tubular components in cells of healthy and tobacco mosaic virus-infected Nicotiana. Virology 33,

Evert, R.F. (1990). Dicotyledons. In Sieve Elements: Comparative Struc- ture, lnduction and Development. H.-D. Behnke and R.D. Sjolund, eds (New York: Springer-Verlag), pp. 103-137.

Gletl, C., and Ziegler, H. (1979). mRNA in phloem exudates. Natur- wissenschaften 66, 523-524.

Klm, S.-G., Chang, J.-R., Cha, H.C., and Lee, K.-W. (1988). Callus growth and plant regeneration in diverse cultivars of cucumber (Cu- cumis sativus L.). Plant Cell, Tissue and Organ Culture 12, 67-74.

Kleinlg, H., Thones, J., Dorr, I., and Kollmann, R. (1975). Filament formation in vitro of sieve tube protein from Cucurbita maxima and Cucurbita pepo. Planta 127, 163-170.

Kollmann, R., Dorr, I., and Kleinig, H. (1970). Protein filaments- Structural components of the phloem exudate. 1. Observations with Cucurbita and Nicotiana. Planta 95, 86-94.

Lackney, V.K. (1991). Ultrastructure and formation of phloem in Cu- curbita pepo callus. J. Struct. Biol. 106, 135-144.

Laemmll,-U.K. (1970). Cleavage of structural proteins during the as- sembly of the head of bacteriophage T4. Nature 227, 680-685.

MacDonald, R.J., Swift, G.H., Przybyla, A.E., and Chlrgwln, J.M. (1987). lsolation of RNA using guanidinium salts. In Methods in En- zymology, Vol. 152, Guide to Molecular Cloning Techniques. S.L.

292-303.

257-283.

26-35.

Berger and A.R. Kimmel, eds (San Diego: Academic Press), pp.

Murashige, T., and Skoog, F. (1962). A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiol. Plant. 15, 473-497.

Read, S.M., and Northcote, D.H. (1983). Subunit structure and inter- actions of the phloem proteins of Cucurbita maxima (pumpkin). Eur. J. Biochem. 134, 561-569.

Sabnls, D.D., and Hart, J.W. (1978). The isolation and some proper- ties of a lectin (haemagglutinin) from Cucurbita phloem exudate. Planta 142, 97-101.

Sabnis, D.D., and Hart, J.W. (1979). Heterogeneity in phloem protein complements from different species: Consequences to hypotheses concerned with P-protein functions. Planta 145, 459-466.

Sambrook, J., Fritsch, G.F., and Maniatis, T. (1989). Molecular Clon- ing: A Laboratory Manual. 2nd ed. (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory).

Sham, M.-H., and Northcote, D.H. (1987). Transcription and transla- tion of phloem protein (PP2) during phloem differentiation in Cucurbib maxima. Planta 170, 392-399.

Smith, L.M., Sabnis, D.D., and Johnson, R.P.C. (1987). Immunochem- ical localisation of phloem lectin from Cucurbita maxima using peroxidase and colloidal-gold labels. Planta 170, 461-470.

Song, C.-S., Yu, J.-H., Bal, D.H., Hester, P.Y., and Klm, K.-H. (1985). Antibodies to the a-subunit of insulin receptor from eggs of im- munized hens. J. Immunol. 135, 3354-3359.

Towbln, H., Staehelln, T., and Gordon, T. (1979). Electrophoretic trans- fer of proteins from polyacrylamide gels to nitrocellulose sheets. Proc. Natl. Acad. Sci. USA 76, 4350-4354.

Walker, T.S. (1972). The purification and some properties of a protein causing gelling in phloem sieve tube exudate from Cucurbita pepo. Biochem. Biophys. Acta 257, 433-444.

Walker, T.S., and Thalne, R. (1971). Proteins and fine structural com- ponents in exudate from sieve tubes in Cucurbitapepo stems. Ann.

Weber, C., Franke, W.W., and Kartenbeck, J. (1974). Structure and biochemistry of phloem-proteins isolated from Cucurbita maxima. Exp. Cell Res. 87, 79-106.

219-227.

BOt. 35, 773-790.

DOI 10.1105/tpc.4.12.1539 1992;4;1539-1548Plant Cell

D E Bostwick, J M Dannenhoffer, M I Skaggs, R M Lister, B A Larkins and G A ThompsonPumpkin phloem lectin genes are specifically expressed in companion cells.

 This information is current as of December 25, 2015

 

Permissions X

https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists