Plant Molecular Biology 17: 19-27, 1991. © 1991 Kluwer Academic Publishers. Printed in Belgium.

19

Expression of inducible angiosperm promoters in a gymnosperm, Picea glauca (white spruce) D . D . Ellis, 1 D. McCabe, 2 D. Russell, 2 B. Martinell 2 and B.H. McCown l

1Department of Horticulture, University of Wisconsin-Madison, Madison, WI 53706, USA; 2Agracetus lnc., Middleton, 1411, USA Received 9 October 1990; accepted in revised form 18 February 1991

Key words: conifer, gene expression, heterologous promoters, inducible promoter activity, particle acceleration, transient assay

Abstract

Electrical discharge particle acceleration was used to test the transient expression of numerous inducible angiosperm promoters in a gymnosperm Picea glauca (white spruce). Promoter expression was assayed in three different tissues capable of in vitro regeneration, zygotic embryos, seedlings and embryogenic callus. The promoters tested include the light-inducible Arabidopsis and soybean ribulose-l,5-bisphosphate small subunit promoters and a maize phosphoenolpyruvate carboxylase promoter; a soybean heat-shock-inducible promoter, a soybean auxin inducible promoter and a maize alcohol dehydrogenase promoter. Promoters were cloned into a promoter-less expression vector to form a promoter-fl-glucuronidase-nopaline synthase 3' fusion. A similar construct was made using the cauliflower mosaic virus 35S (CaMV 35S) promoter as a control. All promoters were expressed in white spruce embryos, yet at levels lower than CaMV 35S. In addition, in the embryos the heat-shock and the alcohol dehydrogenase promoters showed inducible expression when given the proper induction stimulus. In seedlings, expression of all promoters was lower than in the embryos and expression was only inducible with the heat-shock promoter in the cotyledons. Of the tissues tested, the expression level of all promoters was lowest in embryogenic callus. Interestingly, the expression of the fl-glucuronidase gene in embryogenic callus was restricted to the proembryonal head cells regardless of the promoter used. These results clearly demonstrate the use of particle bombardment to test the transient expression of heterologous promoters in organized tissue and the expression of angiosperm promoters in a gymnosperm.

The genetic manipulation through traditional breeding of long-lived perennial crops such as forest trees has been limited, mostly due to the time required until sexual maturity is reached in commercially important timber species. The use of genetic engineering to improve selected superior genotypes could therefore have a significant impact on the rate of genetic improvement of these species. Unfortunately, the successful re-

generation of forest trees containing introduced genes is confined to only a couple of hardwood species, Populus [9, 22, 26, 27, 30] and Juglans [23]. Within the largest and economically most important group of temperate forest trees, the conifers, progress has been made with the identification of Agrobacterium strains infectious to a wide range ofgenotypes [7, 8, 25, 29] and with the use of electroporation or PEG for the intro-

20 duction of foreign D N A into protoplasts [3, 10, 31, 37]. Nevertheless, there have been no reports of the regeneration of transformed conifer plants. One reason is the inability to readily regenerate plants from those cells which contain the foreign DNA. The use of particle acceleration for the introduction of genes into plant cells offers the opportunity to selectively target D N A directly to diverse cell types in organized tissues. Published reports of plant tissue exposed to particle acceleration and transiently or stably expressing introduced genes includes suspension or protoplast-derived cells [14, 15, 22, 24], leaves [15], embryos [6, 4], meristems [19], nodules [22], and pollen [33]. The potential precision offered by microprojectile bombardment for the targeting of D N A to specific cells or cellular components has been demonstrated with the incorporation of foreign D N A into yeast mitochondria [ 12] and Chlamydomonas chloroplasts [5]. The long-term success of introducing genes of commercial interest into trees, as well as other crops, may weigh heavily on the ability to selectively control the expression of the genes. To date, genes introduced into conifers have been driven by promoters characterized as high-expression constitutive promoters, CaMV 35S [3, 8, 10, 37] and Agrobacterium T-DNA promoters [7 ], which may or may not offer some tissue-specific gene regulation. The ability to use well characterized promoter sequences from other plants to regulate gene expression in a tissue- or cell-specific manner in a long-lived perennial, such as conifers, would be desirable. To date, there is no evidence that such regulatory sequences from angiosperms will even function in gymnosperms or that the putative trans-acting factors required for inducible gene expression are present to properly regulate these inducible promoters. In this study, we demonstrate the use of electrical discharge particle acceleration, to introduce genes into gymnosperm tissues and to test the inducible expression of heterologous promoter activity in several different target systems capable of regeneration.

M a t e r i a l s and methods

Open pollinated Picea glauca (white spruce) seed was collected from the Wisconsin Department of Natural Resources seed orchard at Lake Tomahawk, WI. Prior to use, seed was imbibed overnight in water, surface-sterilized for 20 minutes in 20~o chlorox (5.25~ sodium hypochlorite) and rinsed in sterile distilled water. The embryos were aseptically excised and placed on woody plant medium (WPM) [17]. For the seedling experiments, embryos were placed on hormone-free W P M for seven days to allow radicle, hypocotyl and cotyledon elongation prior to particle acceleration. For the embryo experiments, embryos were placed on bud induction medium (WPM supplemented with 50/~M zeatin and 0.01 # M thiadiazuron) for seven days prior to particle acceleration. This pre-treatment has been shown to be necessary for a high level of transient gene expression for white spruce embryos (D.D. Ellis, unpublished). Embryogenic cultures were induced and maintained as previously described [36]. Targets for particle acceleration consisted of fifteen seedlings, twenty embryos or 200 mg fresh weight of embryogenic callus placed on 10 ml of the appropriate medium supplemented with 250 #g/ml carbenicillin in 15 m m × 60 m m disposable Petri plates. For particle acceleration, 0.5 #g of plasmid D N A was adhered to 1 mg of 1-3/~m gold particles by CaCI2 and spermidine precipitation [14]. Bombardment was done at 16 kV with a bead load of 0.05 mg/cm 2 [6, 19]. Constructs to test the expression and inducibility of angiosperm promoters in white spruce were made by fusing the various promoters in front of the fl-glucuronidase (GUS) gene in a pUC19-derived expression vector (Fig. 1). The heterologous promoters used included the light inducible promoters from maize phosphoenylpyryvate carboxylase (PEP) (W. Taylor, unpublished), and Arabidopsis ribulose-l,5-bisphosphate carboxylase IA (rbcS) [32], as well as a soybean hs6871 heat-shock-inducible promoter [28]. These promoters were cloned by polymerase chain reaction (PCR) using oligonucleotide primers synthesized from the published se-

21 500O

~°~

~

.

I/

-'~.~

,

3' fusion was used as a non-inducible control to determine if the induction stimuli used has a general affect on gene expression. The induction protocol to test inducibility of gene expression included: Exposure of the tissue to light for 24 hours prior to G U S assay for the light-inducible promoters (rbcS and PEP); flooding the tissue with sterile distilled water for 24 hours prior to GUS assay for the A D H promoter; - e x p o s i n g the tissue to 42 °C for 1 hour, followed by 3 hours at 27 °C prior to the GUS assay for the heat-shock promoter; either flooding the tissue with water containing 25 #M 2,4-D or transferring the tissue onto WPM medium supplemented with 50 #M 2,4-D for 24 hours prior to GUS assay for the auxin-inducible promoter. The expression of the GUS enzyme in all tissues containing the various promoters was histochemically assayed with 5-bromo-4-chloro-3indolyl glucuronide (x-gluc) [11 ]. All x-gluc assays were done five days after particle acceleration in an attempt to minimize inducible expression which may have been associated with particle bombardment, yet prior to a decrease in maximal G U S expression which occurs after seven days (D.D. Ellis, unpublished). The tissue was placed in x-gluc buffer for 24 hours and then cleared with lactophenol. Blue spots indicating G U S gene expression (Fig. 2) were counted under a dissecting scope to provide a relative level of promoter expression. All treatments with embryos consisted of at least 10 embryos and all -

4000

-

/ 3000

I

Xraol[-,~EO)

-

NCO~ {~O0~

ExampleofpUC19 expressionvector used. Cmc2128 is the soybean heat-shock-GUS-NOS 3' construct.

Fig. 1.

quences such that the resulting amplified fragments contained the entire promoter regions required for proper gene expression (Table 1). The light-inducible soybean SRS 1 rbc S promoter was isolated from pSRS2.1 [4]. These light- and heatshock-inducible promoter sequences were cloned into the pUC 19-derived promoter-less expression vector to form a promoter-GUS-nopaline synthase (nos) 3' gene fusion. The maize alcohol dehydrogenase (ADH) promoter-GUS-nos 3' plasmid was provided by Walker e t al. [34]. The soybean auxin-inducible promoter-GUS plasmid, designated pCB832, was provided by C.S. Brown and T.J. Guilfoyle [20]. The plasmid pCMC1100 [ 19] containing a CaMV 35S promoter-GUS-nos Table 1.

Promoters, sources and their size used to test the inducible expression of heterologous promoters in white spruce.

Inducible promoter

Auxin Heat-shock ADH rbcS rbcS PEP

Promoter-GUSconstructs Source

Reference

Promoter size

Soybean Soybean Maize Soybean

McClure et al. [20] Schoffl et al. [28] Walker et al. [34] Berry-Lower et al. [4] Timko et al. [32] W. Taylor, unpublished

- 830 to -420 to - 1090 to - 1550 to - 320 to - 990 to

Aradopsis

Maize

+1 + 100 + 106 + 45 +1 + 60

22 experiments were replicated three times. With the seedlings, all treatments consisted of at least 10 seedlings and all experiments were replicated twice. A minimum of five embryogenic calli were assayed per treatment. Quantification of G U S enzyme activity was done using 4-methyl umbelliferyl glucuronide (MUG) [11] and was used to confirm that the number of cells histochemicaUy expressing the G U S gene in embryos was related to enzyme activity. Two embryos were pooled per extraction and all data points represent the average of at least six embryos. Total protein quantification from the embryos was done by the method of Lowry et aL [ 18].

Results

Embryos

Using the histochemical x-gluc staining, the expression of the CaMV 35S-GUS construct in embryos was highly uniform in the number of G U S expressing spots (Fig. 2). Spots which represented a larger than one cell area were probably due to a single cell in the spot expressing GUS, yet the monomeric form of the cleaved substrate leaked into surrounding cells prior to oxidative dimerization [11]. With a shorter staining time (3-6 h) faint single-celled blue spots were observed (Fig. 3), yet these were significantly

Fig. 3. White spruce embryo exposed to x-gluc for 4 h showing well defined single-cell expression of GUS (arrows).

harder to count due to the faint blue color and therefore a 24 h x-gluc exposure time was used. A mean of more than thirty G U S spots per embryo were routinely counted when the CaMV 35S-GUS construct was used (Fig. 4). The expression of G U S was evenly distributed over the hypocotyls and cotyledons but expression was absent in the roots. The presence of an outer dead cell mass around the non-elongated root caused by the proliferation of the root cap without radicle elongation (Fig. 2) inhibited particle penetration and thus may have prevented transformation of the cells underneath. The overall CaMV 35SG U S expression in the embryos was not signifi~, 45

~

Ill

+ lndu~lon

4035.

a. 3 0 -

ili

~ 25D~ 20. o 15~6

10.

5"

• o

Fig. 2. White spruce embryos exposed to x-gluc for 24 h showing transient G U S expression (blue spots) 48 hours after particle acceleration.

~

5"

o

~

oo

Fig. 4. The relative level of transient expression of heterologous promoters expressed as the number of GUS spots per white spruce embryo. Bars represent + SE.

23 cantly influenced by any of the induction stimuli used for the other promoters. Although the expression of the CaMV 35S-GUS construct was the only control promoter used, this suggests that the induction stimulus alone does not affect general non-target promoter activity. All angiosperm promoters included in this study were able to drive G U S expression in white spruce embryos, although at levels lower than that of CaMV 35S. No light inducible expression was observed with P E P or either rbcS promoter. This could have been due to a general lack of transient light inducibility, as has been observed with both soybean and cotton (D. Russell, unpublished). The auxin-inducible promoter-GUS construct was also non-inducible by flooding the embryos with an auxin-containing solution or by the placement of the embryos on a high-auxin-containing agar-solidified medium. It is possible that the endogenous level of auxin present in the 7-day-old embryos was sufficient to induce maximal promoter activity and therefore no additional induction would be observed. This auxin-inducible construct has yielded auxin-inducible transient expression in soybean, though the basal level can vary significantly perhaps due to changes in endogenous auxin levels. Both the A D H and the heat-shock promoters showed an increase in G U S expression with induction. In the case of A D H , the overall level of expression was relatively weak and even with flooding, the expression of G U S from this promoter was not significantly different from either the non-inducible heat-shock promoter or the soybean rbcS promoter. In the case of the heatshock promoter, a 5-fold or more increase in G U S expression was routinely observed after a heat-shock treatment. In one replication, the level of induced expression was 10-fold greater than the non-induced control and slightly higher than CaMV 35S (Fig. 5b). The activity of the G U S enzyme was quantified using the fluorometric M U G assay. Figure 5 shows that the relative level of G U S enzyme activity is correlated to the relative number of G U S spots counted per embryo. As expected, the total protein content in different embryos was similar

o

E ,m =

60 5550454035-

~

25-

0

2015-

Z

I I I I + Induction 1::::3 - ;nclucCdon

5 0

heot shock

110

(3-n* ~*.t • -¢=mr3s=

100'

O-PEP

+ light

- light

O

90.

•-P~

..Q

80.

rl-~ •-Ms

b E

PEP

CoMV35s

/I /

I

. LIJ

* Ho*t = * o k - ~ = =~*=k

/ ~-

70 60 50.

/ /

//

/ "

/1

I

-- "

/

I'I'P" / / ,~

30.

I

..7. /

/

/ d

/ Jl~

/ / . ~":~" . ~:~:~--'IIIF

~,~

t"

10,

o= 0

1

2

3

4

5

6

MUG exposure time (hours)

Fig. 5. Comparisonof the relative levels of promoter activity as detected by x-gluc (a, top) and MUG (b, bottom). Bars represent + SE.

and hence relative levels of G U S expression would be the same whether represented on a per embryo or per unit protein basis. A high endogenous background activity caused initial difficulty with the M U G assays and required that the M U G exposure time be extended over five hours. This is in contrast to x-gluc detection of G U S activity where no background G U S activity was detected. In addition, due to the relatively low G U S activity in a transient assay, two embryos were pooled per extraction to insure a detectable level of G U S activity in each sample.

Seedlings

For analysis of tissue-specific G U S expression, the seedlings were divided into roots, hypocotyls and cotyledons. In general, G U S expression in

24 15 in :,,.~--

roots I"--1 hypocotyl cotyledons

"1o ¢1

~100 a.

0

.0 E

Z

~.~.~

~.

.

+

I

+

I

~

~

O

+

+

+

X

I

Fig. 6. The relative levels of transient expression of heterologous promoters expressed as the number of G U S expressing spots per white spruce seedling part. Bars represent _+ SE.

the roots was infrequent, if at all (Fig. 6). Expression in both the hypocotyls and cotyledons was the same with most promoters although the PEP promoter had an elevated level of expression in the hypocotyls while the auxin-inducible promoter had an elevated level of G U S expression in the cotyledons. As with the embryos, there was no significant effect of the induction stimulus on the expression of the CaMV 35S-GUS construct. The only promoter which gave inducible expression in the seedlings was the heat-shock promoter and in this case G U S expression was inducible only in the cotyledons. Overall, the intensity of blue in the G U S expressing spots in the seedlings was less than that observed in the embryos.

types are present in the callus, including elongate suspensor cells attached to the head cells, yet G U S expression was rarely observed in these other cell types. Within the embryogenic head cells, G U S expression was present in all head cells of a G U S expressing pro-embryo. This is most likely due to one or a few cells expressing G U S and diffusion of the monomeric x-gluc cleavage product prior to dimerization throughout the other pro-embryonic head cells.

Embryogenic callus The level of G U S expression in embryogenic callus was highly variable and very low with all promoters except CaMV 35S. No inducible expression was observed, even with the heat-shock promoter. One interesting similarity in the expression of all promoters was that the expression of G U S was confined almost exclusively to the proembryonal head cells (Fig. 7). Several other cell

Fig. 7. A white spruce proembryo showing CaMV 35S-GUS expression confined exclusively to the proembryonal head cells. Note no G U S expression was histochemically detected in the other cell types such as suspensor cells.

25 Discussion

Microprojectiles have been used to introduce D N A into many different plant cells [6, 14, 15, 16, 19, 22, 24, 33, 35 ]. In most of these reports, callus, cell suspensions or pollen were used to study transient expression of marker genes. In this paper, we have shown that electrical discharge particle acceleration can be used as a rapid and efficient method to study the transient expression of promoter activity in intact organized tissue or whole organs The importance of extending the use of microprojectile bombardment to whole organized tissues is that tissue-specific and/or developmentally regulated promoter expression can be tested transiently. Microprojectile bombardment has been previously used to show proper tissue-specific expression of homologous promoters in a transient assay; complementation of anthocyanin mutants in maize was done by introducing intact genes and monitoring for anthocyanin biosynthesis [13]. In animals, myotubulespecific promoters have been shown to transiently express marker genes specifically in myotubules and not myoblasts (S. Johnton, personal communication). These studies clearly demonstrate that genes delivered by microprojectiles into organized tissue can be expressed in a fashion similar to the endogenous gene. With most crop species, tissue culture systems for the regeneration of plants from single cells do not exist. Therefore, in vitro manipulation is confined to differentiation from organized tissue. Such is the case with conifers, where regeneration in tissue culture is mostly limited to organized embryogenic tissues. Therefore, the use of organized tissue for gene expression studies allows for the analysis of parameters which affect transient expression in those tissues that are amenable to regeneration into plants. Because of this, the optimization of methods for studying transient gene expression could also aid in the development of stable gene transfer protocols within this important group of plants. It is clear that the heterologous angiosperm promoters used in this study can be expressed in the gymnosperm, white spruce. This was not the

case with all the heterologous promoters tested, as an auxin-inducible promoter [ 1] and an ATPase AHA-3 promoter (DeWitt et al., unpublished) were not expressed (data not shown). This demonstrates that neither the gold particles or the plasmid DNA caused any background x-gluc detectable expression of G U S activity. The lack of inducible enhancement of gene expression of the promoters tested does not infer that these promoters will not be inducible in vivo in conifers, because most of these promoters were also not inducible in a transient assay in angiosperms. These promoters do however display inducible and/or the expected tissue specific expression in angiosperm transgenic plants (D. Russell, unpublished). It is interesting however that the expression of some of these promoters (auxin, HS, and Arabidopsis rbcS) was at a relatively high level in white spruce embryos. Although the intensity of blue in the tissues was not the same for all the promoters, a decreased intensity was always correlated with a decrease in the number of G U S spots. This supports the use of this technique to determine relative heterologous promoter activity in different tissues and suggests that there may be a threshold amount of the G U S enzyme needed for histochemical detection. With the embryos, further support was provided by the M U G assay which gives the same relative order of promoter strength as did the histochemical assay. Interestingly, the relative intensity of blue staining also correlated with the mean number of blue spots on the different explants, with the highest blue intensity in the embryos, then the seedlings and the lowest intensity in the embryogenic callus. The lower level of expression in the seedlings and embryogenic callus relative to embryos could be attributed in part to the fact that the particle acceleration parameters used were optimized for transient expression in embryos. These parameters are finely tuneable and they must be optimized for each tissue. Both the root tissue and the embryogenic callus have a mucilaginous-like covering requiring a greater discharge voltage to penetrate into the cells. In the case of embryogenic callus, we have found that an increase in dis-

26 charge voltage is required to optimized transient expression. Therefore, the decrease in transient expression in these tissues may reflect a lack of the particles penetrating the cells as much as tissue-specific expression. The inducibility of both the maize ADH and the soybean heat-shock promoters in embryos demonstrates that cis-acting elements from angiosperms will function in gymnosperms. In the case of the heat shock promoter where expression was inducible in both embryos and seedlings, this also indicates that white spruce has the requisite putative trans-acting factors necessary for the regulation of this promoter, confirming the evolutionary conservation of this system. In the seedings however, the heat shock promoter was only inducible in the cotyledons. The lack of inducibility or expression of the ADH promoter in the seedlings, and in particular the roots, could be due to the problem of optimized particle acceleration parameters as previously mentioned. While this reasoning may explain the lack of expression in the roots it clearly does not explain the low expression in both the hypocotyl and cotyledons because expression was observed by the other promoters in these tissues. The data suggests that maize ADH is only weakly, if at all, expressed in white spruce, and that both seedling and embryogenic callus are not competent to respond to anaerobic stress in the required fashion to induce expression. The observation that the auxin-inducible promoter showed a high level of expression without induction in the cotyledons is intriguing. This promoter was originally isolated from soybean hypocotyl tissue and was not strongly expressed in other soybean seedling tissues [21]. The high expression in white spruce embryos could be due to an endogenous auxin level in the seedling tissue which is high enough to allow maximum gene induction, without the addition of exogenous auxin. The fact that the cotyledons are still actively elongating may indicate that they too have a high endogenous auxin supply. By using electrical discharge particle acceleration, we have developed a highly reproducible system where, in a relatively short period of time,

large numbers of cells in an organized tissue can be transformed, and will express foreign DNA. By counting the number of GU S-expressing spots in a given tissue, we have been able to test the relative strength of numerous promoter constructs relatively easily and rapidly. When regeneration of transformed plants is accomplished, the in vivo testing of these and other tissue specific and developmentally regulated heterologous promoters in gymnosperms can proceed.

Acknowledgements The authors thank Trant Marty, Wisconsin Department of Natural Resources for supplying the seed used in this study, Jim Sellmer for his critical review of the manuscript and Jenny Whatever for assistance in the preparation of the manuscript.

References 1. Ainley WM, Walker JC, Nagao RT, Key JL: Sequence and characterization of two auxin-regulated genes from soybean. J Biol Chem 263:10658-10666 (1988). 2. Bekkaoui F, Pilon M, Laine E, Raju DSS, Crosby WL, Dunstan DI: Transient gene expression in electroporated Picea glauca protoplasts. Plant Cell Rep 7: 481-484 (1988). 3. Bekkaoui F, Datla RSS, Pilon M, Tautorus TE, Crosby WL, Dunstan DI: The effects of promoter on transient expression in conifer cell lines. Theor Appl Genet 79: 353-359 (1990). 4. Berry-Lowe SL, Mc Knight TD, Shah DM, Meagher RB: The nucleotide sequence, expression, and evolution of one member of a multigene family encoding the small subunit of ribulose-1,5-bisphosphate carboxylase in soybean. J Mol Appl Genet 1:483-498 (1982). 5. Boynton JE, Gillham NW, Harris EH, Hosler JP, Johnson AM, Jones AR, Randolph-Anderson BL, Robertson D, Klein TM, Shark KB, Sanford JC: Chloroplast transformation in Chlamydomonas with high velocity microprojectiles. Science 240:1534-1538 (1988). 6. Christou P, McCable DE, Swain WF: Stable transformation of soybean callus by DNA-coated gold particles. Plant Physiol 87:671-674 (1988). 7. Dandekar AM, Gupta PK, Durzan DJ, KnaufV: Transformation and foreign gene expression in micropropated Douglas-fir (Pseudotsuga menziesii). Bio/Technology 5: 587-590 (1987). 8. Ellis D, Roberts D, Sutton B, Lazaroff W, Webb D, Flinn B: Transformation of white spruce and other

27 conifer species by Agrobacterium tumefaciens. Plant Cell Rep 8:16-20 (1989). 9. Fillatti JJ, Sellmer J, McCown B, Haissig B, Comai L: Agrobacterium mediated transformation and regeneration of Populus. Mol Gen Genet 206: 192-199 ( 1987). 10. Gupta PK, Dandekar AM, Durzan DJ: Somatic proembryo formation and transient expression of a luciferase gene in Douglas fir and lobolly pine protoplasts. Plant Science 583:85-92 (1988). 11. Jefferson RA: Assaying chimeric genes in plants: the GUS gene fusion system. Plant Mol Biol Rep 5:387-405 (1987). 12. Johnston SA, Anziano PQ, Shark K, Sanford JC, Butow RA: Mitochondrial transformation in yeast by bombardment with microprojectiles. Science 240:1538-1541 (1988). 13. Klein TM, Roth BA, Fromm ME: Regulation of anthocyanin biosynthetic genes introduced into intact maize tissues by microprojectiles. Proc Natl Acad Sci USA 86: 6681-6685 (1989). 14. Klein TM, Gradziel T, Fromm ME, Sanford JC: Factors ~lnfluencing gene delivery into Zea mays cells by highvelocity microprojectiles. Bio/Technology 6:559-563 (1988). 15. Klein TM, Harper EC, Svab Z, Sanford JC, Fromm ME, Maliga P: Stable genetic transformation of intact Nicotiana cells by the particle bombardment process. Proc Natl Acad Sci USA 85:8502-8505 (1988). 16. Klein TM, Wolf ED, Wu R, Sanford JC: High-velocity microprojectiles for delivering nucleic acids into living cells. Nature 327:70-73 (1987). 17. Lloyd GH, McCown BH: Commercially feasible micropropagation of mountain laurel (Kalmia lattfolia) by use of shoot-tip culture. Proc Int Plant Crop Soc 30:421-437 (1980). 18. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ: Protein measurement with the folin phenol reagent. J Biol Chem 193:265-275 (1951). 19. McCabe DE, Swain WF, Martinell BJ, Christou P: Stable transformation of soybean (Glycine max) by particle acceleration. Bio/Technology 6:923-926 (1988). 20. McClure BA, Hagen G, Brown CS, Gee MA, Guilfoyle TJ: Transcription, organization, and sequence of an auxin-regulated gene cluster in soybean. Plant Cell 1: 229-239 (1989). 21. McClure B, Guilfoyle T: Characterization of a class of small auxin-inducible soybean polyadenylated RNAs. Plant Mol Biol 9:611-623 (1987). 22. McCown BH, McCabe DE, Russell DR, Robinson D J, Barton KA, Raffa KF: Stable transformation of Populus and incorporation of pest resistance by electrical discharge particle acceleration. Plant Cell Rep 9:590-594 (1991). 23. McGranahan GH, Leslie CA, Uratsu SL, Martin LA, Dandekar AM:Agrobacterium-mediated transformation of walnut somatic embryos and regeneration of transgenic plants. Bio/Technology 6:800-804 (1988). 24. Mendel RR, Muller B, Schulze J, Kolesnikov V, Zelenin

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

A: Delivery of foreign genes to intact barley cells by high-velocity microprojectiles. Theor Appl Genet 78: 31-34 (1989). Morris JW, Castle LA, Morris RO: Efficacy of different Agrobacterium tumufaciens strains in transformation of pinaceous gymnosperms. Physiol Mol Plant Path 34: 451-461 (1989). Pythoud F, Sinkar VP, Nester EW, Gordon MP: Increased virulence ofAgrobacterium rhizogenes conferred by the vir region of pTiBo542: application to genetic engineering of poplar. Bio/Technology 5:1323-1327 (1987). Riemenschneider DE, Haissig BE: Producing herbicide tolerant poplars using Agrobacterium turnefaciens-mediated genetic transformation: a summary of recent research. In: Ahuja MR (ed) Woody Plant Biotechnology, IUFRO Proceedings (in press). Schoffi F, Raschke E, Nagao RT: The DNA sequence analysis of soybean heat-shock genes and identification of possible regulatory promoter elements. EMBO J 3: 2491-2497 (1984). SederoffR, Stomp A-M, Chilton WS, Moore LW: Gene transfer into loblolly pine by Agrobacterium tumefaciens. Bio/Technology 4:647-649 (1986). Sellmer JC, McCown BH: IX Transformation in Populus spp. In: Bajaj YPS (ed) Plant Protoplasts and Genetic Engineering II, pp. 155-172. Springer-Verlag, Berlin Heidelberg (1989). Tautorus TE, Bekkaoui F, Pilon M, Datla RSS, Crosby WL, Fowke LC, Dunstan DI: Factors affecting transient gene expression in electroporated black spruce (Picea mariana) and jack pine (Pinus banksiana) protoplasts. Theor Appl Genet 78:531-536 (1989). Timko MP, Herdies L, de Almeida E, Cashmore AR, Leemans J, Krebbers E: Genetic engineering of nuclearencoded components of the photosynthetic apparatus in Arabidopsis. In: The Impact of Chemistry on Biotechnology. ACS Symp Serv 362:279-295 (1988). Twell D, Klein TM, Fromm ME, McCormick S: Transient expression of chimeric genes developed into pollen by microprojectile bombardment. Plant Physiol 91:1270-1274 (1989). Walker JC, Howard EA, Dennis ES, Peacock WJ: DNA sequences required for anaerobic expression of the maize alcohol dehydrogenase 1 gene. Proc Natl Acad Sci USA 84:6624-6628 (1987). Wang Y-C, Klein TM, Fromm M, Cao J, Sanford JC, Wu R: Transient expression of foreign genes in rice, wheat and soybean cells following particle bombardment. Plant Mol Biol 11:433-439 (1988). Webb DT, Webster F, Flinn BS, Roberts DR, Ellis DD: Factors influencing the induction of embryogenic and caulogenic callus from embryos of Picea glauca and P. engelmanff. Can J For Res 19:1303-1308 (1989). Wilson S, Thorpe T, Maloney M: PEG-mediated expression of GUS and CAT genes in protoplasts from embryogenic suspension cultures ofPicea glauca. Plant Cell Rep 7:704-707 (1989).