Molecular and Cellular Biochemistry 97: 145-151, 1990. © 1990 Kluwer Academic Publishers. Printed in the Netherlands.

Original Article

Preparation and characterization of monoclonal antibodies to human hexokinase type I

Laura Chiarantini, Giordano Serafini, Vilberto Stocchi and Mauro Magnani Istituto di Chimica Biologica 'G. Fornaini', Universit~ degli Studi, Via Saffi, 2 - 61029 - Urbino, Italy

Received 26 January 1990; accepted 16 March 1990

Key words: hexokinase, monoclonal antibodies, comparative studies

Abstract

1. Three different immunization protocols and several screening procedures were used to prepare seven mouse monoclonal antibodies to human placenta hexokinase type I. None of these monoclonals were able to recognize the native enzyme but all detected hexokinase when adsorbed onto polystyrene plates or on immunoblots after SDS/polyacrylamide-gel electrophoresis. 2. All seven monoclonals recognize the two different subtypes of human hexokinase I equally well. Limited tryptic digestion of hexokinase followed by Western blotting and immunodetection show that these monoclonals recognize epitopes that lie in different tryptic peptides. 3. Comparative ELISA studies showed that human hexokinase types I and II have great immunological similarities while hexokinase I from different mammalian species and yeast hexokinase are recognized with different affinities.

Introduction

Hexokinase (HK; EC 2.7.1.1) is one of the rate limiting enzymes of glycolysis catalyzing the phos- phorylation of glucose to glucose 6-phosphate with Mg ATP as the phosphate donor. Four distinct isoenzymes [1, 2] are usually present in varying proportions in mammalian tissues [3-5]. Human placenta contain mainly HK type I and this was found to be present in two subtypes [6] that differ in molecular weight (Mr 112,000 and 103,000 respec- tively). Monospecific polyclonal antibodies raised in rabbits against high and low molecular weight hexokinases detected both subtypes on immuno- blots [6]. As one approach towards identifying the significance of the presence of hexokinase I sub- types and to possibly separate them from each other, we developed a library of monoclonal anti- bodies directed against this enzyme. In this report we describe the results obtained and the useful

application of such antibodies to perform compara- tive studies between HK isoenzymes and among the same isoenzyme in different species.

Materials and methods

Materials

Coenzymes, enzymes and substrates were obtained from Sigma Chemical Co. or Boehringer Biochem- ia. Activated carboxyhexyl-CH-Sepharose 4B and Protein A Sepharose CL 4B were from Pharmacia, DE-52 (DEAE-cellulose) was from Whatman. Matrex Gel Red A, Matrex Gel Blue A and ultra- filtration membranes (PM 30) were from Amicon. Electrophoresis and Western Blotting equipment and reagents were from Bio-Rad Laboratories.

Assays Hexokinase activity was measured as described in

tOE c

OJ

d

OJ U

d

0 U/

q- 0

- o

50

146

5 iO 20 40 80 160 3E8 640 1280 2560

An±ibody Di[utlon

Fig. 1. ELISA Titration curves of monoclonal antibodies against human hexokinase I. Human placenta hexokinase was adsorbed onto polystyrene plates at 0.1/zg/well. The antibodies used were 2El lC6 (O); 2B7 (--); 19Dll (A); 25E10 (0); 24B3 (×); 26E (+); 17D10 ( I ) . The first five monoclonals are IgM and the last two IgG. The ELISA procedure and antibody purifications are described under 'Materials and methods'. The results are means of four determinations that agreed within 5%.

[6], protein was determined by the method of Brad- ford [7].

Purification of hexokinase Hexokinase type I from human placenta was puri- fied as described in [6] to a specific activity of 190 + 5 units/mg of protein. Human hexokinase type II was used after the DE-52 column chromatography step [6]. Hexokinase I from rat brain was purified by the method described in [8] with the exception that after the DEAE chromatography step the en- zyme solution was precipitated by ammonium sul-

fate at 70% saturation, dialyzed and chromato- graphed on Sepharose-N-aminohexanoylglucosa- mine as described in [6] and finally further purified on Matrex Gel Blue A with elution by glucose 6-phosphate [8]. Pig and cow hexokinase I were obtained from heart tissues after ion-exchange chromatography as for the human placenta en- zyme.

Immunization and cell fusion Homogeneous hexokinase type I from human pla- centa was used to immunize 4-6 week old male Balb/c mice by using three different protocols.

Protocol I This is essentially as reported in [9] and utilizing 115/zg of protein for 16 days.

Protocol 2 50/zg of protein was emulsified in 50/zl of MPL + TDM (Monophosphoryl Lipid A + Trehalose Dimycolate from Ribi Immuno- Chem Research, Inc.) adjuvant and injected under the foot pads of each mouse, 13 days later the mice were again inoculated with a similar amount of enzyme and adjuvant.

Protocol 3 Essentially as in [10] 300/zg of protein were used for 32 days.

Three days after the last booster the mice were killed by cervical dislocation and in cases of proto- col 1 and 3 the spleens were removed aseptically, while using protocol 2 the lymphnodes of the knees were utilized, which were well visible after immu- nization. Spleen cells or lymphocytes (1 - 5 × 10 7) were fused with hypoxanthine-guanine phosphori- bosyl transferase negative SP2/0 Ag 14 mice myelo- ma cells in the ratio 10 : 1 as previously described [11] with some modifications as reported in [12]. Hybridoma cells producing antibodies to human

Table 1. Immunization and type of screening utilized for the production of monoclonal antibodies against human placenta hexokinase

Protocol Type of immunization Duration Procedure of screening Positive Class Monoclonals (days) clones

1 Intraperitoneal 16 El isa- Goat 4 IgM 2 Intravenous anti Mouse IgG-Perox

16 Elisa - Goat 10 IgM 10 anti Mouse IgG-Perox

32 Elisa-Protein A-Perox 57

Subplantar

Intraperitoneal IgM 5 IgG2b 1 IgG3 1

147

Fig. 2. Recognition of human hexokinase (5/xg/line) after SDS/polyacrylamide gel electrophoresis and Western Blotting. As first antibodies we used 2EllC6 (1); 2B7 (2); 19Dll (3); 25E10 (4); 24B3 (5); 26E (6); 17D10 (7). As second antibody a goat anti mouse Ig peroxidase conjugate was used. Colour development was obtained by 4-cloro-1 naptol and HzO2. The first line is a Coomassie blu stained gel.

placenta were cloned using the method of limiting dilution [13]. Positive clones were propagated both for the production of antibodies or for storage un- der liquid nitrogen in 10% dimethylsulfoxide, 90% foetal calf serum (FCS).

Different screening procedures were used to de- tect antibodies against HK in the culture medium.

Screening techniques ELISA. At the beginning an ELISA (Enzyme Linked Immunosorbent Assay) was employed us- ing the method reported in [11] with the following modification: the antigen was coated onto 96 well plates at 0.1/zg/well in sodium borate buffer 50 mM, pH 8.2, overnight 4 ° C. As second antibod- ies we used peroxidase conjugated goat antimouse IgG in order to select the clones producing immu- noglobulins (Ig) from the G class as soon as pos-

sible. As an alternative to anti mouse IgG we also used a peroxidase conjugated Protein A.

Western blotting. Electrophoretic procedures and immunoblotting were described in a previous pa- per [6] with the difference that as first antibody we used the growing medium of our hybrid cell lines. In some experiments native hexokinase was first digested with trypsin (1/xg for every 50 tzg of HK) for 15 min at room temperature. The sample was then boiled in SDS buffer and electrophoresed as above; western blotting of digests were performed to identify the HK peptides recognized by each antibody.

Concentration and~or purification of the immunoglobulins The medium coming from hybrid cell lines produc- ing immunoglobulins of the IgM class was con-

148

Fig. 3. Recognition of human hexokinase I subtypes by monoclonal antibodies. The experiment was as in the legend to Fig. 2 except that the amount of hexokinase was 3.5 txg (line 1), 2/xg (line 2); 1 tzg (line 3); 0.7 ~g (line 4); 0.35/.~g (line 5).

centrated by ammonium sulfate fractionation (2EllC6, 19Dll, 2B7, 25E10, 24B3). The ammo- nium sulfate salt was slowly added to the medium containing the immunoglobulins to achieve a 45% saturation. The suspension was gently stirred for 30 min and then centrifuged for 60 min at 16,000 g. The supernatant was removed and retested for the residual activity by ELISA. The pellet was re- suspended in 1/20 of the initial volume in PBS (150 mM NaCI, 10 mM phosphate pH 7.0) and then dialyzed overnight in the cold room against the same buffer.

The concentrated medium was stored at 4 ° C for several months without any loss of activity and used diluted for several experiments reported below.

The hybrid cell lines producing IgG (26E; 17D10) were grown in a medium without FCS and with a substitute of it (SP Nutridoma, Boeringher Mannheim) at the final concentration of 1% v/v. The IgG were purified from the medium by chro- matography on immobilized Protein A and elution by sodium citrate 0.1 M pH 3.5.

Results

At first we tried to generate monoclonal antibodies against human HK by short term immunization followed by spleen cell fusion and screening with

goat anti mouse immunoglobulin conjugated with peroxidase) (Protocol 1). By this method only Ig's of the M class were generated. To increase the possibility of having Ig's of the G class, that are more useful for a number of studies (because of stability, facility of purification etc.) we tried a subplantar immunization protocol (Protocol 2). Again only IgM's were obtained. Finally we immu- nized the mice by intraperitoneal antigen adminis- trations followed by a screening procedure that employed Protein A conjugate to peroxidase as detecting reagent instead of goat anti mouse immu- noglobulin. Since Protein A is more specific for immunoglobulins of the IgG class we expected some positive results in terms of monoclonal anti- bodies of the IgG class. As shown in Table 1 we finally obtained five IgM and two IgG monoclonal antibodies against human hexokinase I. These have been identified as: 2Ell , C6, 2B7, 19Dll, 25E10, 24B3, 26E, 17D10 and used in further stud- ies. Tritation curves performed using the ELISA method for each monoclonal were done to evaluate the optimal dilution at which the concentrated or purified monoclonal could be used. The results obtained are shown in Fig. 1. As regards 24B3 it has a very low response by ELISA technique but, as shown below, it works quite well in HK detection by western blotting.

It is well known that the binding of monoclonal

antibodies to many proteins can be strongly influ- enced by enzyme protein conformation changes.

Bearing this in mind, we studied the ability of our monoclonals to recognize HK in a different confor- mation state (i.e. when the enzyme is in solution or denaturated after boiling in SDS and mercaptoeth- anol). Pure HK 6 mU was incubated for 30 min at 37 ° C with increasing amounts of monoclonal anti- bodies (from 0 to 3/xg). After this incubation time the activity was assayed to see if some of the mono- clonals had direct effect on the catalytic properties of HK. None were found able to inactivate the HK activity. Subsequently we added 50/xl of a 10% suspension of Staphylococcus aureus cells to the same incubations. In case the monoclonal was of the IgM class, the S. aureus was conjugated to a rabbit anti mouse IgM as described in [14]. After a further 30 min incubation at 37°C the suspensions were centrifuged in an Eppendorf microcentrifuge and the residual hexokinase activity assayed in the supernatant. None of our monoclonals seem to recognize the native enzyme is solution independ- ently of the presence of enzyme substrates or inhib- itors (not shown). This is further proof that mono- clonals can react in a different way depending on the method they are tested [15]. To obtain further conformation change human hexokinase was com- pletely denaturated by boiling in the presence of SDS and mercaptoethanol and then separated by SDS/polyacrylamide gel electrophoresis. As shown in Fig. 2 our monoclonal worked very well in the western blotting detecting the enzyme in the dena- tured state.

Recognition of hexokinase I subtypes As previously published [6], the human placenta HK type I shows two components that are con- stantly seen in SDS gel electrophoresis. These two forms do not differ in their isoelectric points, but they have different molecular weights. As shown above, all our monoclonals recognized both the high and low molecular weight subtypes of human hexokinase I. To search for possible differences in hexokinase subtype affinities, we loaded SDS/po- lyacrylamide gels with decreasing amounts of hexo- kinase (from 3.5 to 0.35~g/well) and performed western blotting experiments as above. No differ-

149

Fig. 4. Peptide mapping of human hexokinase by monoclonal antibodies. Human hexokinase was partially digested by in- cubation under native conditions in the presence of trypsin (1/zg for every 50 ~g of hexokinase) at room temperature for 15 min. The sample was then boiled in the presence of SDS 0.5% (w/v) and mercaptoethanol 2% (v/v) and then submitted SDS/po- lyacrylamide gel electrophoresis and Western Blotting. The intracellulose was than incubated with monoclonals: 2B7 (line 1), 17D10 (line 2), 2E11 C6 (line 3), 24B3 (line 4), 19Dll (line 5). The first line on the left represents a Coomassie blue stained gel of the trypsin digested enzyme.

ence among these monoclonals in their ability to recognize the two forms was detected (Fig. 3).

Peptide mapping To identify the hexokinase peptides recognized by the monoclonals studied we performed a limited tryptic digestion of native hexokinase I. The gener- ated peptides were then fractionated by SDS/po- lyacrylamide gel electrophoresis and transfered to

150

d

U d

E 0

e4_

121.

-p

Z

0 -p

-p u

Ill

£

0".

m 0

~2

q- 0

,M

3

4

i £

IOC

50_

0 2E11 C6

5

m

!

1

!

i

l I i

I , I 297 191311 25EI0 24133 26E 17DIO

Monoc[ono,[ An±lbody

Fig. 5. Comparative cross reactivities of anti human hexokinase I monoclonal antibodies against human hexokinase type II (column 1); rat brain hexokinase I (column 2), pig heart hexokinase I (column 3); cow heart hexokinase I (column 4); yeast hexokinase (column 5). Each hexokinase was adsorbed onto polystirene plates at 0.1/zg/well and the absorbance at 492 mm obtained by the ELISA technique compared to the values found for the human placenta hexokinase I (100%). All values are means of 10 different experiments that agreed within 5%.

nitrocellulose membranes by Western blotting. The monoclonals utilized were found to recognize epitopes present in different peptides (Fig. 4).

Comparative studies The reactivity of the seven monoclonal antibodies described above was also tested against human hexokinase type II and hexokinase I from different mammalian species as well as against the yeast enzyme.

By a comparative ELISA we obtained the results reported in Fig. 5. HK type II is well recognized by all antibodies. 17D10 shows a lower affinity for HK I from rat brain and cow heart and at the same time it is the monoclonal that has the highest affinity for yeast HK. 24B3 shows a lower affinity for HK I from cow and pig heart. The yeast protein is poorly recognized by them.

Discussion

The results reported in this paper suggest that hu- man hexokinase is a poor antigen since immuniza- tion protocols previously utilized successfully for the preparation of monoclonal antibodies provided only a few IgM's. One possible explanation for this behaviour can be the high homology between the human and rodent hexokinases [16-18].

To overcome this problem we used different strategies changing both immunization protocols (including the adjuvants) and screening proce- dures. We finally obtained some positive results in terms of monoclonals of the IgG class and several IgM's. The extensive use of these monoclonals proves that they do not recognize human hexoki- nase under native conditions but are quite useful in Western Blotting and comparative studies. In fact, we have been able to show that a very high immun- ological similarity exists between human hexoki-

nase type I and II. Furthermore, in confirmation of other results previously reported by us [19] and other laboratories [16] great similarities also exist among other mammalian hexokinases I. The aim of the present study was also the identification of possible differences among hexokinase I subtypes. Unfortunately, none of the antibodies reported in this paper were able to distinguish among HK sub- types, further suggesting that these show very few differences and that strategies other than immun- ologically ones should be used to characterize them.

Acknowledgements

This work was supported by C.E.E. grant BAP 00551, C.N.R. and M.P.I.

References

1. Katzen HM, Schimke RT: Multiple forms of hexokinase in the rat: tissue distribution, age dependency and properties. Proc Natl Acad Sci USA 54: 1218-1225, 1965

2. Gonzales C, Ureta T, Sanchez R, Niemeyer H: Multiple molecular forms of ATP: Hexose 6-phosphotransferose from rat liver. Biochem Biophys Res Comm 16: 347-352, 1964

3. Colowich SP: The Hexokinases. Enzymes 3rd ed 9: 1-48, 1973

4. Purich DL, Fromm HJ, Rudolf FB: The hexokinases: ki- netic, physical and regulatory properties. A Meister (Eds) Advances in Enzymology, vo139 pp 24%326 (1973) Wiley, New York

5. Wilson JE: Regulation of mammalian hexokinase activity. R Beitner (eds) Regulation of Carbohydrate metabolism, vol 1 48-85, CRC press, BOKA

6. Magnani M, Stocchi V, Serafini G, Chiarantini L, Fornaini G: Purification properties and evidence for two subtypes of human placenta hexokinase type I. Arch Biochem Biophys 260: 388-399, 1988

7. Bradford MM: A rapid and sensitive method for the quanti- tation of microgram quantities of protein utilizing the prin-

151

ciple of protein. Dye binding. Anal Biochem 72: 248-254, 1976

8. Polakis PG, Wilson JE: Proteolytic dissection of rat Brain hexokinase; Determination of the cleavage pattern during limited digestion with trypsin. Arch Biochem Biophys 234: 341-352, 1984

9. Cianfriglia M, Armellini D, Massone A, Mariani M: Simple immunization protocol for high frequency production of soluble antigen-specific hybridomas. Hybridoma 2: 451- 468, 1983

10. Finney KG, Messer JL, De Witt DL, Wilson JE: Mono- clonal antibodies against rat brain hexokinase. J Biol Chem 259: 8232-8237, 1984

11. Shulman M, Wilde CD, K6hler G: A better cell line for making hybridomas secreting specific antibodies. Nature (London) 276: 269-270, 1978

12. De Witt DL, Smith WL: Purification of prostacyclin syn- thase from bovine aorta by immunoaffinity cromatography. J Biol Chem 258: 3285-3393, 1983

13. Legkovits I, Waldmann H: Limiting dilution analysis of cells in the immune system. Cambridge University Press, London and New York, 1979

14. De Witt DL, Rollins TE, Day JS, Ganger JA, Smith WL: Orientation of the active site and antigenic determinations of prostaglandin endoperoxide synthase with endoplasmic reticulum. J Biol Chem 256: 10375-10382, 1981

15. Hollander Z, Katriz EK: Use of monoclonal antibodies to detect conformational alterations in lactate dehydrogenase isoenzymes 5 on heat denaturation and on adsorption to polystyrene plates. Mol Immunology 23: 927-933, 1986

16. Ureta T, Smith AD, Wilson JE: Hexokinase a from mam- malian brain: comparative peptide mapping and immun- ological studies with monoclonal antibodies. Arch Biochem Biophys 246: 41%427, 1986

17. Sehwrab DA, Wilson JE: Complete amino acid sequence of rat brain hexokinase, deduced from the cloned cDNA, and proposed structure of a mammalian hexokinase. Proc Natl Acad Sci 86: 2563-2567, 1989

18. Nishi S, Seino S, Bell GI: Human hexokinase; sequences of amino-acid and carboxyl-terminal halves are homologous. Biochem Biophys Res Comm 157: 937-943, 1988

19. Magnani M, Serafini G, Chiarantini L, Stocchi V: Simi- larities and differences between human and rat hexokinase type I. Mol Cell Biochem 1989 (in press)

Address for offprints." M. Magnani, Instituto di Chimica Biolog- ica 'G. Fornaini', Universit~ degli Studi, Via Saffi, 2 - 61029 - Urbino, Italy