Reporter Transgene Design

When designing a transgenic reporter construct, careful planning is of the utmost importance. The production and analysis of transgenic animals is both labor- and time-intensive, and a well-designed construct will give you a good base on which to plan your subsequent experiments. Although the exact layout of a given reporter depends on the aims of the experiment in question, certain general considerations should be addressed. These will be discussed below.

3.1.1. General Considerations

1. A reporter transgene requires the capacity to work like any other gene. It must contain all of the structural elements recognized by the transcriptional, posttran-scriptional, and translational machinery of the host, leading to the production of a functional reporter gene product.

The primary goal of an experiment aimed at identifying control sequences responsible for particular aspects of gene expression during development is to reproduce the exact pattern of expression of the endogenous gene. This provides a baseline from which sequences controlling this expression can be defined. If there is no previous knowledge about regulatory elements, it is best to begin working with the largest available fragment of DNA, thus maximizing the chances of encompassing all of the required control sequences. Cloning strategies that employ artificial chromosome vectors derived from yeast (YAC), bacteria (BAC), or bacteriophage P1 (PAC) enable the sequences of very large (>100 kb) genomic regions to be tested (10-12).

2. A well-designed vector will prove to be a great help both for the initial cloning of your reporter and its subsequent analysis. Some plasmid vectors, such as pPolyIII (13), have been especially designed with extensive multiple cloning sites for the cloning and excision of transgenic constructs. It may, however, be advantageous to engineer your own.

3. The reporter gene fragment needs to be purified away from vector sequences prior to microinjection (see Note 1). For this reason, the construct should be flanked by unique restriction enzyme sites. The most commonly used enzyme for this purpose is NotI (8-bp recognition sequence), which is an infrequent cutter within eukaryotic DNA. Other suitable enzymes are now commercially available, such as AscI, PacI, PmeI, SfiI, and FseI (8-bp recognition sequences, New England Biolabs) or the extremely rare cutting intron-encoded endonucleases (e.g., I-SceI, 18-bp recognition sequence, Boehringer).

4. Try to utilize a vector that does not contain the lacZa region, since we have found that problems with high frequencies of recombination may arise when trying to insert the lacZ gene into such a vector (see Note 2).

5. Certain eukaryotic DNA sequences may cause problems during cloning, giving rise to a high frequency of recombination events. If this is the case, try switching to a lower copy number cloning vector, or switch to a different bacterial host strain, since this may circumvent the problem (see Note 3).

6. When developing transgenic experiments to define regulatory elements, you will undoubtedly wish to make deletions of your basic construct. It is useful to include suitable restriction sites that will allow the generation of 5'- and 3'-nested deletions using exonuclease III (14). At later stages of the analysis, you may wish to introduce point mutations into specific ds-regulatory elements. It is a good idea to utilize a vector that will allow the employment of site-directed mutagenesis without the need to subclone specific fragments (15).

3.1.2. Intact Gene Constructs

The usual starting point for investigations into gene regulation requires the structural organization of the transgene to be as close as possible to that of the endogenous gene, thus allowing the context of regulatory elements to remain unchanged. For this reason, an intact gene construct is normally used in which the reporter is inserted, in frame, into the coding sequence of the gene leading to the production of a fusion protein. Alternatively, it is possible to create a bicistronic transgene, whereby translational initiation of the reporter is mediated by a viral internal ribosome entry site (IRES) sequence (see Note 4). It is desirable to insert the reporter either at, or as close as possible to, the transla-tional initiation codon (ATG). This is particularly important if your gene product has known biological activity, since overexpression of the fusion protein may interfere with the regulation of your transgene or have more widespread biological effects in the transgenic organism. Such constructs should contain:

1. 5'-Flanking sequences, including 5'-untranslated regions (5'-UTR), and should at least encompass the minimal elements capable of promoter activity (usually located proximally to the transcription start site) enabling the production of mRNA with a functional translational initiation codon (AUG). More distal 5'-sequences (sometimes many kilobases upstream) may also contain important regulatory regions.

2. 3'-Sequences containing elements that specify correct transcriptional termination and the addition of poly(A) to the 3'-end of the mRNA (see Note 5). The 3'-untranslated regions (3'-UTR) of certain genes include sequences that confer rapid degradation of the encoded mRNA and may have to be removed to obtain satisfactory levels of transgene activity (see Note 6). 3'-Transcriptional regulatory regions may also be present.

3. Intragenic regions: Introns have been shown to increase the overall levels of transgene activity (16,17) and, in many cases, contain transcriptional regulatory elements. However, there is no absolute requirement for splicing for the production of a functional reporter gene product. The intronic sequences of some genes are not required for appropriate transcriptional regulation and may, in some cases, be omitted. For example, in the case of the myogenin gene (see Fig. 1), only upstream flanking sequences appear to be necessary for appropriate transcrip-tional regulation (6).

Fig. 1. (A,B) Dynamic progression of myogenin gene expression during the development of the skeletal musculature. The embryos shown, stained with X-gal for P-galactosidase activity, are from a transgenic line carrying a nuclear localized lacZ reporter under the control of 1.1 kb of myogenin 5'-flanking sequence (Ashby and P. W. J. R., unpublished data; see ref. 6). At 11.5 dpc (A) intense bars of staining can be seen within the myotomal component of the somitic mesoderm. By 13.5 dpc (B), specific muscle blocks are clearly distinct. (C) Use of dual reporter transgenes within the same embryo. A coronal section through the hindbrain of a 9.5-dpc embryo is shown (anterior is uppermost). Blue staining resulting from P-galactosidase activity is derived from expression of the lacZ gene under the control of a Hoxb-2 enhancer, which is active in rhombomeres 3 and 5. A second construct harbors the alkaline phosphatase gene under the control of an enhancer from Hoxb-2, which directs expression in rhombomere 4 (revealed as brown staining). Photographs were kindly provided by N. Collins and B. Singh (myogenin) and J. Sharpe (Hoxb-2), Division of Developmental Neurobiology, N.I.M.R, Mill Hill. (See color plate 9 appearing after p. 368.)

Fig. 1. (A,B) Dynamic progression of myogenin gene expression during the development of the skeletal musculature. The embryos shown, stained with X-gal for P-galactosidase activity, are from a transgenic line carrying a nuclear localized lacZ reporter under the control of 1.1 kb of myogenin 5'-flanking sequence (Ashby and P. W. J. R., unpublished data; see ref. 6). At 11.5 dpc (A) intense bars of staining can be seen within the myotomal component of the somitic mesoderm. By 13.5 dpc (B), specific muscle blocks are clearly distinct. (C) Use of dual reporter transgenes within the same embryo. A coronal section through the hindbrain of a 9.5-dpc embryo is shown (anterior is uppermost). Blue staining resulting from P-galactosidase activity is derived from expression of the lacZ gene under the control of a Hoxb-2 enhancer, which is active in rhombomeres 3 and 5. A second construct harbors the alkaline phosphatase gene under the control of an enhancer from Hoxb-2, which directs expression in rhombomere 4 (revealed as brown staining). Photographs were kindly provided by N. Collins and B. Singh (myogenin) and J. Sharpe (Hoxb-2), Division of Developmental Neurobiology, N.I.M.R, Mill Hill. (See color plate 9 appearing after p. 368.)

3.1.3. Basic Promoter Constructs

1. Once specific regulatory regions have been defined, they may then be tested individually on a simpler, basic promoter construct. Sequences can be tested in both 5'- and 3'-contexts and in both orientations. This will provide information on what subset(s) of the endogenous gene's expression pattern is specifiable in the absence of other elements and test for enhancer function. The basic structure of this type of construct consists of a minimal promoter (see Note 7) linked to a reporter gene, e.g., lacZ/SV-40p(A), to which putative regulatory sequences can be apposed. The most widely used promoters for this purpose are listed in Table 1.

2. Negative regulatory regions can be tested on a construct that can direct ubiquitous or widespread expression to see if they are capable of restricting this pattern. Various heterologous promoters have been identified that may be suitable for

Table 1

Some Heterologous Transgene Promoters

Name

Source

Size, bp

Notes

Refs.

Minimal promoters

P-Globina Human P-globin 48

Hsp68a

Hoxb-4

HMGCR

Mouse heat-shock gene (hsp70.1)

Mouse Hoxb-4 gene

HSV-TK

Ubiquitous promoters P-actin

Herpes simplex virus thymidine kinase gene (-105 to +51)

Chicken P-actin gene

Mouse hydroxy methylglutaryl CoA reductase gene

Mouse Hoxa-7 gene

12CC

550C

159C

Weak; contains 6,18

TATA box and initiator element; clean Strong; no activity 5,19,20 in embryo in absense of enhancer elements6 Contains two 5,21

independent initiators; expression only in the superior coliculi in the midbrain Ectopic expression 22-25 only

Ubiquitous; larger, 1,26,27 intron containing, human promoter gives stronger expression Ubiquitous embryonic 28,29 expression; contains untranslated exon and intron

Uniform expression 30 throughout mid-gestation embryo aSee Note 3S. bSee Note 39.

this purpose (see Table 1). It is probably more desirable though to test such regions on a construct containing the homologous promoter.

Table 2

Reporter Genes and Substrates

Reporter

Cellular localization

Form

Suitability Substrates

Refs.

ß-Galactosidase

Cytoplasm Nucleus

Cytoplasmic membrane

Axonally transported

Cytoplasmic membrane (extracellular)

Native form SV-40 T-Ag nuclear localization signal fused to lacZ Synthetic transmembrane domain fused to lacZ Tau (microtubule associated protein) fused to lacZ Native form

All genes All genes

Transmembrane or secreted

Axonally expressed

All genes BCIP/NBT

and double staining with lacZ

31,49,50

X-gal (light blue or Bluo-gal (dark blue)

32-34

(blue-purple), BM purple (deep purple), NATP/fast red RC (red), or NAGP/fast blue BN (green)

3.1.4. Choice of Reporter Gene

1. Various reporter systems are available, the choice of which depends largely on the particular experimental aims and the cell types in which the construct is expressed (see Table 2). LacZ reporters can be targeted to the cytoplasm, the nucleus, and the cytoskeleton, or anchored to the cell membrane (31). Two chro-mogenic substrates are available for use with lacZ reporters, X-gal, and Bluo-gal (see Subheading 3.4.1.).

2. Although lacZ is often the reporter gene of choice, another option is the AP system, which utilizes the human placental alkaline phosphatase gene (PALP-1: 32-34). PALP-1 has a number of attributes that make it useful as a transgenic reporter: The cDNA is small (approx 2 kb as opposed to 3.2 kb for lacZ) and is readily expressed in mammalian cells. In addition, the human AP protein is 100-fold more resistant to heat treatment than that of the mouse, and there are several inhibitors of mouse AP activity that do not affect the human enzyme (33). A range of substrates is available for AP that give different colored products (see Table 2). Its main advan-

tage is that it may be used in conjunction with lacZ to produce "double transgenics," where the expression of two reporter genes can be compared in the same embryo (see Note 8 and Fig. 1). Another potentially useful application is in the study of the transcriptional regulation of two linked genes. In this case, lacZ can then be used to monitor the expression of one gene and PALP-I that of the other, allowing the analysis of elements that may act on one or both genes.

3. The green fluorescent protein (GFP) from the jellyfish Aequorea victoria has received widespread interest as a reporter owing to its potential for the direct real-time imaging of gene expression and protein localization in vivo (35-37). GFP has been successfully employed as a reporter within a range of organisms (38,39), although its application to mammalian species has been hindered by problems with both levels of protein expression and intensity of fluorescence. These problems may, in part, be due to thermal effects and the requirement for molecular oxygen for the formation of the GFP chromophore (40,41). The continued development of variant GFPs, which have altered spectral properties and improved transcriptional, translational, and posttranslational characteristics, lends great hope for the success of this versatile marker gene in mammalian systems (41-48). A useful forum for the discussion of GFP and related topics is the Fluorescent Proteins Newsgroup, available via USENET at bionet.molbio. proteins.fluorescent or via the World Wide Web at http://www.bio.net/hypermail/ FLUORESCENT-PROTEINS/.

Was this article helpful?

0 0

Post a comment