Alkaline phosphatase as a reporter enzyme

11

Gene, 66 (1988) 11-17 Elsevier GEN 02400

Alkaline phosphatase as a reporter enzyme (Spectrophotometric

assay; histochemical

assay; transfection

efficiency)

Kyonggeun Yoon, Mark A. Thiede and Gideon A. Rodan Department of Bone Biology and OsteoporosisResearch, Merck Sharp & Dohme Research Laboratories, West Point, PA 19486 (U.S.A.) Received Revised

28 May 1987 9 February

1988

Accepted

12 February

Received

by publisher

1988 25 February

1988

SUMMARY

This study examines the use of alkaline phosphatase (AP) as a reporter enzyme. We constructed a plasmid containing the cDNA which encodes the bone/liver/kidney rat AP under the control of the simian virus 40 (SV40) early promoter and used it to transfect Chinese hamster ovary, SV40-transformed African Green Monkey kidney 7, and rat osteosarcoma 25/l mammalian cells. AP activity in these cells, measured three days later, was 40-400-fold above background. When AP and chloramphenicol acetyltransferase (CAT) plasmids were cotransfected, the detection of AP activity by a simple spectrophotometric assay was at least as sensitive as the detection of CAT activity using a radioactive substrate. Moreover, since mammalian AP is a membranebound ectoenzyme, transfected cells can be visualized by histochemical staining. This approach was used to estimate transfection efficiency. The convenient methods for AP detection should make it a useful reporter enzyme.

INTRODUCTION

Transfection of DNA into eukaryotic cells is a powerful tool for identifying sequences required for

Correspondence

to: Dr. K. Yoon, Department

Merck Sharp & Dohme Research 19486 (U.S.A.)

of Bone Biology,

Laboratories,

West Point, PA

Tel. (215)661-7823.

the regulation of gene expression. In such studies a putative promoter is joined to a reporter gene, which frequently codes for an enzyme. The enzymatic activity is then assayed to estimate the level of gene

de; EtdBr, ethidium (see MATERIALS globulin;

Abbreviations:

A, absorbance;

Ab, antibody;

phatase;

Ap, ampicillin;

albumin;

CAT, Cm acetyltransferase;

bp, base pair(s);

ovary;

Cm, chloramphenicol;

Green

Monkey

Dulbecco’s

kidney;

modified

037%1119/88/$03.50

AP, alkaline phosBSA, bovine

CHO,

Chinese

COS, SV40-transformed

DEAE,

diethylaminoethyl;

Eagle’s medium;

0 1988 Elsevier

DMSO,

serum

AND METHODS,

of replication;

P, promoter;

hamster

PolIk, Klenow (large) fragment

African

resistance;

DMEM,

thin-layer

B.V. (Biomedical

FACS, fluorescence

activated

Division)

p-nitrophenyl

phos-

on’, origin phosphate;

of E. coli DNA polymerase

ROS, rat osteosarcoma; chromatography.

AS-MX

acid 2,4-dimethylanilide; PNPP,

cell

salt solution

section b); Ig, immuno-

kb, 1000 bp; LB, Luria broth; naphthol

phate, 3-hydroxy-2-naphthoic

dimethylsulfoxi-

Science Publishers

bromide;

sorter; FCS, fetal calf serum; HBS, Hanks balanced

I; R,

SV40, simian virus 40; TLC,

12

Sma

I

Xba I

sac I Eco RI

EcoRl

I

Kpn

Fig. 1. Construction

and structure

I

of AP plasmids.

(A) Structure

of the AP expression

kb) was digested with EcoRI + BglII and blunted with PolIk. This EcoRI-BglII was ligated to the plasmid comprised (stippled

pECE (Ellis et al., 1986) previously

152-bp untranslated

5’-sequence

(hatched

DNA carrying

is located

pSV2AP.

AP cDNA

(2.4 and

The AP cDNA in this construct

(filled box) and 325-bp untranslated

SV40 early promoter

237 bp upstream

A full-length

(2 kb) was purified by gel electrophoresis

with SmaI and phosphatased.

box), 1572-bp coding sequence

box) of rat AP cDNA. The open box represents

(ori). The major cap site in the SV40 early promoter

digested

vector,

fragment

3’-sequence

(Pn) and the SV40 origin of replication

from the first AUG codon of AP cDNA.

The open

13

expression.

Enzymes

Transformants

which have been used for this and Escherichia coli

purpose include CAT, h&erase

Ap/ml.

were grown in LB containing

Small-scale

plasmid

DNAs

100 fig

were prepared

Ow et al.,

(Holmes

et al., 1980) and digested with several re-

1986; deWet et al., 1987; An et al., 1982). The desirable properties of a reporter enzyme include

striction inserted

enzymes to confirm the orientation of the DNA. For transfection experiments, plas-

high sensitivity,

mid DNA was isolated by the modified alkaline lysis method, followed by CsCl-EtdBr equilibrium density

/&galactosidase

reliable

(Gorman

et al.,

low background

1982;

and a convenient,

assay.

Alkaline hydrolyzes

gradient

phosphatase is a stable enzyme which a wide range of substrates (orthophos-

phoric monoesters) spectrophotometric

centrifugation

(Maniatis

et al., 1987) were kindly provided by Drs. W.J. Rutter at University of California, San Francisco,

at alkaline pH. Several simple methods have been developed to

quantitate AP activity, using both absorbance and fluorescence measurements (McComb et al., 1979).

and S. Subramani Diego. Restriction

Moreover, since AP is a membrane-bound ectoenzyme, cells expressing AP can be visualized by histochemical staining. To explore the potential use of AP as a reporter enzyme, we cloned the full-length cDNA coding for bone/liver/kidney isoenzyme of rat AP (Thiede et al., 1988) and placed it under the control of the SV40 early promoter. This AP plasmid was transfected into several mammalian cells. Both total enzymatic activity and transfection efficiency were measured in transfection experiments, using convenient assays of AP. The sensitivity of AP detection compared favorably with that of CAT detection in cells cotransfected with AP and CAT plasmids.

and PolIk were purchased Mannheim (Indianapolis, IN).

MATERIALS

(b) Transfection

of the promoterless

AP vector,

construct.

two polyadenylation

Instead,

transcription fragment

from pBR322.

containing

DNA sequences from pBR322 promoters.

represents

pSVOAP. The promoter addition

the full length coding sequence

and enhancer

of pSV2AP.

activity

measured

and tissue cultures

the polyadenylation

from pBR322

sequences

were inserted

fragment

encoding

addition

site (&)

the ApR(AmpR)

of SV40 (Pn and ori of pSV2AP)

at the 5’ side of the AP cDNA

of pSVOAL (deWet

et al., 1987) was ligated

and the SV40

gene. (B) Structure were deleted in this

to prevent

any cryptic

to a 2-kb HindIII-Sac1

As for pSVZAP, the open box at the 3’ end of the AP cDNA represents

and the SV40 polyadenylation

the ApR gene. Several unique restriction

The AP-specific

containing

the sequence

sites (A,)

The 4.3-kb HindIII-Sac1

of the SV40 small-t intron

encoding

the SV40 sequence

for splicing. The single line indicates

Boehringer-

fection. Cells were washed with DMEM and the medium was replaced with 2 ml of DMEM containing plasmid DNA complexed to 0.2 mg/ml DEAE-dextran (M, 500000; Pharmacia P-L Biochemicals, Piscataway, NJ). Following 1 to 4 h incubation at 37 ‘C, the DNA solution was aspirated and replaced with 2 ml of 10% DMSO in HBS (137 mM NaCl, 5 mM KCl, 0.7 mM Na,HPO,, 6 mM glucose, 21 mM Hepes, pH 7.1) for 90 s. The DMSO solution was then removed, cells were washed and fresh medium containing 10% FCS was added. Following incubation at 37 ‘C for 24, 48 and 72 h, cells were rinsed two times with Ca2+ -free and

The AP expression vector (pSV2AP) and the promoterless vector (pSVOAP) were constructed as shown in Fig. l,A and B. E. coli strain DH5a (B.R.L., Gaithersburg, MD) cells were transformation hosts for these plasmids (Ham&an, 1983).

small-t intron sequence

from

reagents used for tissue culture were obtained from Gibco Laboratories (Grand Island, NY). Transfection was carried out using the DEAE-dextran method (Lopata et al., 1984). Briefly, 1.0 x lo6 cells were plated in a 60-mm dish one day before trans-

of plasmids

4 box shown at the 3’ end of AP cDNA

at University of California, San endonucleases, T4 DNA ligase

CHO and COS7 cells were maintained in DMEM containing 10% FCS. The ROS 25/l cells were maintained in F12 medium containing 10% FCS. All

AND METHODS

(a) Construction

et al., 1982). Plas-

mids pECE (Ellis et al., 1986) and pSVOAL (deWet

addition

site (A,). The single line represents

sites, for example Hind111 and SalI, are available

in CHO cells transfected

with pSVOAP was less than

the sequence

for inserting

1 nmol/min/mg.

various

14

Mg2 + -free

HB S, scraped into 1.O ml of a solution containing 0.04 M Tris - HCl, pH 7.4 and 0.15 M NaCl and pelleted for 2 min in a microfuge at 4°C. Cell extracts were prepared of AP-lysis

I

Time course

of alkaline

phosphatase

mM MgCl, and 0.1 y0 Triton X-100). AP activity in frozen cell extracts was stable for at least two months. and histochemical assays of

alkaline phosphatase

Time (h)

assay

One to 5 % of the total cell extracts were sonicated twice for 5 s and the AP activity was measured in 0.2 ml of an assay buffer containing 0.1 M 2-amino-2methyl-1-propanol (pH 10.5) 2 mM MgCl, and 2 mM PNPP. The mixture was incubated for 3 to 20 min at 37 ’ C and the reaction was stopped by adding 0.8 ml of 1 M NaOH. The absorbance of the reaction mixture was measured at 410 nm and the specific activity of AP was calculated as nmol of PNPP hydrolyzed per min per mg protein (Majeska et al., 1985). The total protein in each cell extract was measured according to Spector (1978) using BSA standards. (2) Histochemical assay For the histochemical staining of AP, cells were washed with 0.15 M NaCl, and were incubated for 30 min with a solution (volume = 10 ml) containing 0.4 ml naphthol AS-MX phosphate (pH 8.6), 9.6 ml 0.15 M NaCl and 2.4 mg Fast Violet B (Ackerman, 1962). After staining, cells were fared with 100% ethanol. All chemicals were purchased from Sigma (St. Louis, MO).

RESULTS

changes

AP specific activity

(nmol/min/mg)b

ROS25/1

CHO

in trans-

AND DISCUSSION

(a) Spectrophotometric measurement of alkaline phosphatase activity in transfected cells CHO, ROS 25/l and COS7 cells were transfected with the AP plasmid pSV2AP, using the DEAEdextran method (see MATERIALS AND METHODS, section b). The endogenous AP activity in these cells ranged from 1 to 5 nmol/min/mg protein

cos7

0

1

24

13

290

22

48

37

430

120

72

130

350

190

1

a Three cell lines, ROS25/1,

(I) Spectrophotometnk

activity

fected cells a

by lysing cells in 100 ~1

buffer (10 mM Tris * HCl, pH 7.5, 0.5

(c) Spectrophotometric

TABLE

with 2 pg of pSV2AP AND METHODS,

CHO and COS7, were transfected

plasmid

as described

section b. Cell extracts

and 72 h after transfection

5

in MATERIALS

were harvested

and AP activities

24,48

were measured

for

each time point. b The spectrophotometric described

l-5 y0 of total cell extract. 57 x A,,,/min/mg Coomassie

assay

in MATERIALS

for AP was

AND METHODS, AP-specific

protein.

Protein

blue, using BSA standards

value represents

carried

activity was calculated was

measured

(Spector,

assays

were run on triplicate

as with

1978). Each

the mean from at least two independent

fection experiments;

out as

section cl, using

trans-

samples,

(Table I). Cells with high endogenous AP activity such as HeLa, kidney cells and osteoblasts would not be suitable hosts for the use of AP as a reporter enzyme. Transfection per se did not turn on expression of endogenous AP since CHO, ROS 25/l and COS7 cells transfected with pSV2CAT or with salmon sperm DNA showed no measurable increase in AP activity (not shown). Following transfection with pSV2AP, the AP activity was 40-400-fold higher than the endogenous levels (Table I) and was well within the measurable range. Of the total cell extract from a 60-mm plate of transfected CHO cells 5% gave a spectrophotometric reading of A 0.58 at 410 nm in the AP assay run for 3 min at 37°C. This corresponds to 100 ng of AP protein per 700 pg total protein in the 60-mm culture dish, and is approx. 2000-fold above the limit of detection of AP by PNPP hydrolysis (Nair et al., 1987). Table I also shows the time course for the expression of AP following the transfection of these cells: in CHO cells the AP activity peaked at 48 h, while in ROS 25/l and COS7 cells AP activity continued to rise for three days. These differences may be due to different turnover rates of AP mRNA or AP protein in these cells.

15

(b) Transfection

efficiency measured by histochemi-

cal staining of alkaline phosphatase

1/

3

I

I

I

I

I

I

2

4

6

0

10

12

Since the catalytic site of AP is extracellular, several histochemical methods can be used to demonstrate AP activity. The AP-catalyzed hydrolysis product of naphthol AS-MX phosphate coupled with a diazonium salt (Fast Violet B salt) forms an insoluble red pigment which precipitates at sites of AP activity on the cell surface. Fig. 3 illustrates the histochemical staining of AP for the three cell lines presented in Table I, 72 h after transfection. This procedure provided a direct measure of the number of cells transfected with pSV2AP. A close correlation was observed between transfection efficiency and AP specific activity in CHO and ROS 25/l cells: a three-fold increase in the transfection efficiency of CHO cells was reflected in a three-fold higher AP specific activity (Fig. 3 and Table I). On the other hand, COS7 cells exhibited a higher APspecific activity than RO S 25/ 1 cells, although fewer COS7 cells were histochemically stained: 7% compared to 12% for ROS 25/l cells. This difference probably reflects efficient replication of transfected plasmids in COS7 cells, which contain large T antigen (Gluzman, 198 1). The dual measurement of total enzymatic activity and transfection efficiency would

Amount of DNA Transfected ($ug)

Fig. 2. Relationship amount

between

of transfected

the AP-specific

pSV2AP

density of 5 x 104/cmZ were transfected trations

ofpSV2AP

described

plasmid

in MATERIALS

extracts

were harvested

AP activity

activity

using the DEAE-dextran AND METHODS,

in MATERIALS

method

section

transfection

experiments.

obtained

after transfection

Maximal

for

METHODS, from two

AP activity

with 4 pg of plasmid

as

b. Cell

and assayed

AND

section cl. Each point is an average of results obtained separate

at the

with a range of concen-

72 h after transfection

as described

and the

DNA. CHO cells plated

was

pSV2AP.

Fig. 2 shows the relationship between AP activity and the amount of pSV2AP DNA used in the transfection of CHO cells. AP specific activity increased with the amount of plasmid transfected up to about 4 pg DNA when it reached a plateau.

Fig. 3. Demonstration demonstrated visualized stained scanned

of AP-positive

cells by histochemistry.

72 h later by histochemical

staining

described

All cells were transfected in MATERIALS

as bright red spots, and their surface areas were quantitated

with a Joyce-Loebl

red was divided by the whole area since the cells were confluent and the average

fraction

of positive

cells was calculated.

with 2 pg of pSV2AP

AND METHODS,

section

Image Analyzer.

and AP activity

d. AP-positive

In each field, the area which

at this stage. At least live different

fields in each plate were

Plates: 1, ROS 25/l; 2, CHO; 3, COS7. The quantitation

12 y0 of ROS 25/l, 30% of CHO and 7% of COS7 cells were transfected

by pSV2AP plasmid, indicating

was

cells were

different transfection

shows that efficiencies.

16

be useful in experiments

designed

strength of a particular promoter differ in transfection efficiency. During

the histochemical

to estimate

the

among cells which

pSV2CAT

plasmids

experiment,

(Fig. 4). In this cotransfection

the AP-specific

activity

was approxi-

mately half of that found in cells transfected

procedure

described

pSV2AP

plasmid

alone, suggesting

with the

an equal uptake

above the cells are fixed and killed. Viable cells transfected with pSV2AP can also be visualized and

of pSV2AP and pSV2CAT by CHO cells (compare Figs. 2 and 4). The maximal expression of AP in

retrieved using immunohistochemical methods. A monoclonal antibody (mAb 143-3) against rat AP was recently utilized to measure the lateral mobility

cotransfected cells was 270 nmol/min/mg total cell protein, corresponding to 1.6 pmol AP enzyme; the maximal

of transfected

total cell protein,

AP in various cell types (Noda et al.,

1987). The transfected cells were visualized rect immunofluorescence using fluorescein cyanate-conjugated

Fab goat anti-mouse

by indiisothioIgG. This

technique can be used for cell sorting, for example with FACS, and for the cloning of viable cells transfected with the AP plasmid. (c) Comparison of alkaline phosphatase and chloramphenicol acetyltransferase reporter enzymes To evaluate the usefulness of AP as a reporter enzyme, CAT and AP plasmids, driven by the same SV40 early promoter, were cotransfected into CHO cells. AP and CAT activities were measured after cotransfection of equal amounts of pSV2AP and

1I

.;

Fig. 4. Comparison transfected

of DNA Transfected

of pSV2AP

after transfection.

AP (open

are presented

plas-

were harvested

72 h

box) and CAT (blackened

box)

as nmol of PNPP

protein and pmol of acetylated

Cm/min/mg

The CAT assays were performed (1982) and the radioactivity TLC plate counting.

was cut out and estimated CAT enzyme (Pharmacia

way, NJ) which acetylates to calibrate

hydrolyzed/min/mg protein,

as described

associated

respectively.

by German

et al.

with each spot on the by liquid

scintillation

P-L Biochemicals,

120 000 nmol of Cm/min/mg

the CAT activity

higher level of AP relative to CAT protein could be due to the higher stability of the AP protein or of the mRNA coding for AP, which were shown to have half lives of 72 h (Nair et al., 1987a) and 10 h (Noda and Rodan, 1987), respectively. It could also be due to more efficient translation of AP-coding mRNA in mammalian cells. The higher abundance of AP seems to compensate for the higher turnover number of CAT, making AP a reporter enzyme of similar sensitivity to CAT. (d) Conclusions The existence of convenient spectrophotometric and histochemical assays for AP, makes this protein a useful reporter enzyme in cells with low endogenous AP activity. Total enzymatic activity and transfection efficiency can be quantitated in transfected cells. The presence of AP on the cell surface can be used- to identify and separate viable transand cell sorting using AP and show that the favorably with

CHO cells were

and pSV2CAT

mids (2, 4, 6 and 8 pg each). Cell extracts activities

to 0.03 pmol CAT

enzyme (Nair et al., 1987a; Shaw et al., 1968;1979). The abundance of AP in cotransfected cells thus appears to be 50-fold higher than that of CAT. The

(pg)

of AP and CAT activities.

with equal amounts

of CAT was 80 pmol/min/mg

corresponding

fected cells using anti-AP antibodies method. Cotransfection experiments CAT plasmids in mammalian cells sensitivity of AP detection compared that of CAT detection.

6

Amount

expression

of CHO cell extracts.

Piscatawas used

ACKNOWLEDGEMENTS

We thank Dr. William J. Rutter for the pECE plasmid and Dr. Suresh Subramani for the pSVOAL plasmid. We acknowledge the advice of Dr. Jang Han for the transfection experiments. We also thank Dr. David Thompson for guidance with the Joyce Loebl Image Analysis system and Mr. Kiki Chang and Mr. Robert F. Buenaga for technical assistance.

17

McComb,

REFERENCES

R.B,, Bowers, G.N. and Posen, S.: Alkaline

tase. Plenum, Ackermann,

G.A.: Substituted

tives for the localization Lab. Invest.

naphthol

AS phosphate

deriva-

of alkaline

phosphatase

activity.

11 (1962) 563-566.

An, G., Hidaka,

in animal

L.: Expression

of bacterial

cells. Mol. Cell. Biol. 2 (1982)

1628-1632. deWet, J.R., Wood, K.V., DeLuca, ramani,

S.: Firefly luciferase

in mammalian

M., Helinski,

gene: structure

D.R. and Suband expression

cells. Mol. Cell. Biol. 7 (1987) 725-737.

Ellis, L., Clauser,

E., Morgan,

Rutter,

W.J.:

residues

1162 and 1163 comprises

activity

and

uptake

of insulin

receptor

tyrosine

insulin-stimulated

of 2-deoxyglucose.

Cell

L.F. and Howard,

mammalian

Y.: SV40 transformed D.S. and Quigley,

preparation

acetyltransferase

simian cells support

plasmids.

the replica-

Anal. Biochem.

DNA

of a CAT

transfection

shock

Sollner-Webb,

treatment.

gene

for the

114 (1981)

Arch.

Biochem.

Biophys.

G.A.: Type /3 transforming

of alkaline

phosphatase

in osteoblastic

rat

and

254

growth

other

osteosarcoma

factor

phenotypic

cells.

J. Cell.

133 (1987) 426-437.

Acid

sulfoxide

Res.

of endogenous

phosphatase

or

12 (1984)

cell line ROS

G.A.: Glucocorticoid

regu-

in the osteoblastic

17/2.8. Endocrinology

Ow, D., Wood,

A Laboratory Spring Harbor,

E.F. and Sambrook,

osteo-

116 (1985)

Manual.

J.: Molecular

Cold Spring Harbor

NY, 1982.

Cloning.

Laboratory,

K.V., DeLuca,

alkaline

phospha-

membrane

protein.

M.R., deWet, J., Helinski,

S.H.: Transient

firefly luciferase

J.

and stable

gene in plant

Cold

expression

cells and transgenic

D.R. of the plants.

Science 234 (1986) 856-859. W.V. and Brodsky,

phenicol

R.F.: Characterization

acetyltransferase

Shaw, W.V., Packman, Hartley,

from

specified

structure

of chloram-

chloramphenicol

resistant

95 (1968) 28-36.

B.D., Burleigh,

B.S.: Primary

A.D., Morris,

H.R. and

of a chloramphenicol

by R plasmids.

Nature

acetyl-

282 (1979)

870-872. Spector, T.: Refinement quantitation. Thiede,

T., Fritsch,

and transfected

tase: a phosphatidylinositol-anchored

Staphylococcus aureus. J. Bacterial.

by DEAE-dextran

170-179. Maniatis,

regulation

transferase

of alkaline

sarcoma

isoenzymes.

Noda, M. and Rodan,

and Howell,

B.: High level

with a dimethyl Nucl.

R.J., Nair, B.C. and Rodan,

lation

placenta

II. Struc-

bone, kidney,

Cell Biol. 105 (1987) 1671-1677.

5707-5716. Majeska,

and

Shaw, D. and

expression

mediated

G.A.: Rat alkaline phosphatase, the osteosarcoma,

18-27.

J.A., Bennet,

between

mobility in

Cell 23 (1981) 175-182.

M.: A rapid boiling method

ofbacterial

M., Cleveland,

glycerol

R.J., Rodkey,

similarities

Physiol.

193-197. transient

D.E., Majeska,

and polyclonal

254 (1987a)

C.D. and Rodan,

markers

cells. Mol. Cell. Biol. 2 (1982) 1044-1051.

tion of early SV40 mutants.

Lopata,

of monoclonal

Biophys.

phos-

of the enzyme

tural

kinase

B.H.: Recombinant

genomes which express chloramphenicol

Holmes,

generation

Arch. Biochem.

Nair, B.C., Johnson,

G.A.: Rat alkaline

and characterization

Noda, M., Yoon, K., Rodan, G.A. and Koppel, D.E.: High lateral

C.M., Moffat,

Gluzman,

R.J. and Rodan,

I. Purification

antibodies.

45 (1986)

721-732. Gorman,

phatase,

Phospha-

1979.

(1987b) 28-34.

D.O., Edery, M., Roth, R.A. and

Replacement

Nair, B.C., Majeska, from osteosarcoma:

K. and Siminovitch,

/3_galactosidase

New York,

G.A.:

M.A., Yoon, The

Communicated

K., Golub,

structure

alkaline phosphatase; Natl. Acad.

ofthe Coomassie

Anal. Biochem. of rat

blue method of protein

86 (1978) 142-145. E.E., Noda, osteosarcoma

the product

(ROS

17/2.8)

of a single copy gene. Proc.

Sci. USA 85 (1988) 319-323. by N. Davidson.

M. and Rodan,