A family with high serum leucine aminopeptidase activity derived from a novel variant CD13

^a Author for correspondence. Fax 81-3-5803-0161;

	Abstract

Top Abstract Introduction Patient and Methods Results Discussion References

 
We investigated a family in which some individuals showed extremely high serum leucine aminopeptidase (LAP) (EC 3.4.11.2) activity, mainly derived from a variant CD13. The isoelectric points of the variant and normal CD13 were 3.3 and 4.1, respectively, and both points converged at 4.4 after treatment with neuraminidase, indicating that more sialic acids are bound to the variant CD13 than normal. The molecular masses of both CD13s were 144 kDa by sodium dodecyl sulfate–polyacrylamide gel electrophoresis. After treatment with neuraminidase, N-glycosidase, and O-glycosidase, apparent molecular masses of the variant and normal CD13 were 106 kDa and 100 kDa, respectively, suggesting that the variant CD13 contains a longer peptide than normal. This is the first case of familial high serum LAP activity in which the origin could be demonstrated by anti-CD13 monoclonal antibodies to be a variant CD13 inherited in an autosomal dominant mode. The isoelectric point of the LAP activity after neuraminidase treatment was different from that previously reported.

	Introduction

Top Abstract Introduction Patient and Methods Results Discussion References

 
Catalytic activity of serum aminopeptidase (EC 3.4.11.2) is an indicator of intra- or extrahepatic biliary obstruction. This enzyme removes N-terminal amino acids from peptides and hydrolyzes artificially synthesized substrates such as L-leucine- or L-alanine-p-nitroanilide. Because of the broad specificity of the substrates, there has been some confusion concerning nomenclature: Serum aminopeptidase is also called leucine aminopeptidase (LAP, the term we use here), alanine aminopeptidase (AAP), aminopeptidase N, aminopeptidase (microsomal), or arylamidase.¹ In addition, two or more aminopeptidases may attack the same substrate. For example, cystine aminopeptidase (EC 3.2.11.3), which increases in sera of pregnant women, also cleaves L-leucine-p-nitroanilide. For this reason, applying specific monoclonal antibodies may be useful in a study to clarify to what extent a particular enzyme contributes to serum LAP activity.

LAP is located in various tissues including the small intestine, renal proximal tubules, and the canalicular domain of hepatic cell membranes. The cDNA of an enzyme with aminopeptidase activity has been cloned and found to be identical to CD13, a cell surface marker of granulocytes, monocytes, and their progenitors (1)(2)(3)(4). The results of a previous study involving an inhibition assay with specific monoclonal anti-CD13 antibodies (3)(5) demonstrated that aminopeptidase activity in sera from healthy individuals and patients with hepatobiliary disorders is predominantly attributable to CD13 (6). No report, however, has demonstrated the contribution of CD13 to familial high serum aminopeptidase activity.

We investigated a family in which three of five members showed extremely high serum LAP activity in spite of their not having any other physical or laboratory abnormality. Using anti-CD13 monoclonal antibodies, we discovered that this activity was mainly due to a novel variant CD13.

Although in the US and in European countries serum aminopeptidase (EC 3.4.11.2) is often called AAP, in our country it is conventionally measured as LAP, with L-leucine-p-nitroanilide as a substrate. Because this example of familial high serum aminopeptidase was discovered by measurement of LAP activity and also by electrophoresis, in which it was detected with L-leucine-ß-naphthylamide, we thus use LAP in this communication.

	Patient and Methods

Top Abstract Introduction Patient and Methods Results Discussion References

 
The propositus, a 34-year-old woman, visited our hospital because her high serum LAP activity had been noted when she went to a gynecological clinic for treatment of endometriosis. She had had no particular disease other than endometriosis and showed no particular abnormality in the physical findings or the routine laboratory examination except for the high serum LAP activity. Chest x-ray film, electrocadiography, abdominal ultrasonography, and abdominal computed tomography revealed no abnormalities. She had no symptom suggesting the immunological dysfunction of leukocytes.

Serum samples were collected from the propositus, her father, mother, paternal grandmother, and younger sister, none of whom had any particular disease, and an unrelated healthy volunteer (a 37-year-old man) as a control. The procedures followed were in accordance with the Helsinki Declaration of 1975, as revised in 1983, and with the ethical standards for research of our university.

measurement of serum lap activity
LAP activity was measured by a Model 736 automated analyzer (Hitachi) with reagents LAP-HR (Wako Pure Chemical Industries): L-Leucine-p-nitroanilide was used as a substrate and absorbances at 415 nm (main) and 505 nm (sub) were monitored to measure the production of p-nitroaniline at 37 °C in 80 mmol/L Tris-HCl buffer, pH 7.4. The enzyme activity was expressed as U/L, where 1 U represents the activity required to generate 1 µmol of p-nitroaniline per minute. The reference interval was 33–73 U/L, and the upper detection limit was 700 U/L. Serum samples that showed higher LAP activities than this limit were remeasured after appropriate dilution.

high-voltage isoelectric focusing on cellulose acetate membranes
Serum, 5 µL, or partially purified CD13 was analyzed by high-voltage isoelectric focusing. This was performed according to a method developed in our laboratory that involves laying a maximum of six sheets of cellulose acetate membranes Separax-EF (Fuji Photo Film) on top of each other for simultaneous electrophoresis, resulting in six copies of the identical electrophoretic pattern (7)(8). In this study, three sheets of cellulose acetate membranes that had absorbed 100 g/L ampholyte Sepaline (pH 3.5–10) (Fuji Photo Film) and 100 g/L sucrose were used simultaneously. After electrophoresis, one sheet was used for protein staining with Coomassie Brilliant Blue G-250, one for LAP activity staining, and one for immunoblotting.

lap-activity staining of cellulose acetate membranes
The cellulose acetate membrane after isoelectric focusing was incubated at 37 °C for 15–30 min in a solution containing 0.55 mmol/L L-leucine-ß-naphthylamide (Sigma), 0.84 mmol/L Fast-Garnet (Sigma), 0.65 mmol/L magnesium sulfate, and 32 mmol/L sodium phosphate (pH 6.6).

immunoblotting with anti-cd13 antibodies
After isoelectric focusing, a sheet of polyvinylidene difluoride (PVDF) membrane Immobilon (Millipore) soaked with Tris-glycine buffer (25 mmol/L Tris, 192 mmol/L glycine) was laid on the cellulose acetate membrane and left for 30 min at room temperature. After blocking with Tris-buffered saline (TBS: 20 mmol/L Tris, 137 mmol/L NaCl, pH 7.6) containing 50 g/L skim milk for 1 h, the PVDF membrane was incubated with 20 mg/L mouse anti-human CD13 monoclonal antibody MCS-2 (Nichirei) or B-F10 (Biosource International) in TBS at room temperature for 1 h and washed three times with TBS containing 1 mL/L Tween-20 for 10 min each time. Next, it was incubated with sheep anti-mouse IgG peroxidase conjugate (Sigma) diluted at 1:1000 in TBS at room temperature for 1 h and again washed three times (10 min each) with TBS containing 1 mL/L Tween-20. Bound antibody was detected with ECL reagents (Amersham).

sodium dodecyl sulfate–polyacrylamide gel electrophoresis (sds-page)
SDS-PAGE was performed on a 75 g/L polyacrylamide separating gel according to Laemmli's procedure (9). Protein staining was carried out with Coomassie Brilliant Blue R-250 or Silver Stain II Kit (Wako Pure Chemical).

partial purification of serum cd13 by deae cellulose ion-exchange column chromatography
Normal serum, 8 mL, was diluted 1:20 with 20 mmol/L piperazine buffer (pH 5.0) to reduce the ionic strength and then loaded on a 3-mL column of DE52 (Whatman BioSystems). After extensive washing with the same buffer, CD13 was eluted with 250 mmol/L NaCl and 20 mmol/L piperazine (pH 5.0), dialyzed against PBS (10 mmol/L sodium phosphate, 150 mmol/L NaCl, pH 7.2) at 4 °C overnight, and concentrated with Minicon-CS15 (Amicon).

immunoprecipitation of serum cd13
Serum, 100 µL, or partially purified CD13 was mixed with 15 µL of MCS-2 (100 mg/L) and incubated for 2 h at 4 °C; then 2 µL of rabbit anti-mouse IgG -globulin (28.7 g/L) was added and the mixture was incubated overnight at 4 °C. After centrifugation at 7000g for 5 min, the precipitate was washed twice with TBS containing 1 mL/L Triton X-100 and once with TBS, then subjected to SDS-PAGE or measurement of LAP activity. Heat-inactivated MCS-2 (98 °C, 1 min) or normal mouse IgG was used as a control. To measure LAP activity in the immunoprecipitate, it was redissolved in 50 mmol/L glycine-HCl buffer, pH 2.5, and applied to the method described above, where pH of the applied sample was adjusted to 7.4 by dilution with LAP-HR buffer solution.

purification of cd13
After immunoprecipitation and SDS-PAGE, CD13 was eluted from the excised gel piece by overnight shaking in 200 µL of 20 mmol/L Tris buffer containing 10 g/L SDS (pH 8.0) at 4 °C. Cold (4 °C) acetone, 1 mL, and 5 µg of glycogen were added to the eluate and left for 1 h at -80 °C. The precipitate was dried and dissolved in 10 µL of 20 mmol/L sodium phosphate, pH 7.4, containing 10 mL/L Nonidet P-40 and 1 g/L SDS.

treatment with neuraminidase, n-glycosidase, and o-glycosidase
One hundred microliters of serum or partially purified CD13 was incubated with 10 µL of neuraminidase (III) (NRH-301, Toyobo) in piperazine-N,N'-bis-(2-ethanesulfonic acid) (PIPES) buffer, pH 6.5 (10 U/L neuraminidase, 10 mmol/L PIPES) for 2 h at 37 °C.

Purified CD13 was treated with N-glycosidase F (0.5 U/L) (Boehringer Mannheim) and (or) bovine serum albumin-free O-glycosidase (0.2 U/L) (Boehringer Mannheim) in 20 mmol/L sodium phosphate buffer, pH 7.4, containing 10 mL/L Nonidet-P40 and 1 g/L SDS, overnight at 37 °C.

	Results

Top Abstract Introduction Patient and Methods Results Discussion References

 
lap activity in serum
Because a serum sample of the propositus diluted at 1:5 with PBS showed 449 U/L of LAP activity, which was within the measuring range (<700 U/L), we determined her serum LAP activity as 2245 U/L, 31 times higher than the normal upper limit (73 U/L). Likewise, the LAP activities of her father and paternal grandmother were determined as 1860 U/L and 1980 U/L, respectively. On the other hand, her mother and younger sister showed normal serum LAP activity (Fig. 1 ).

View larger version (20K): [in this window] [in a new window]   Figure 1. Serum LAP activity of the family. The members represented by closed symbols had abnormally high LAP activity; those represented by open symbols had LAP activity in the reference interval. Squares represent men; circles represent women. , dead, LAP activity not known.

immunoprecipitation of serum cd13
Serum samples of the propositus containing 54.4 mU of LAP activity each were subjected to immunoprecipitation with an anti-CD13 monoclonal antibody MCS-2 or a control IgG. As shown in Table 1 , 71% of the activity was removed from the serum and 45% of the removed activity was recovered in the precipitate. In contrast, heat-inactivated MCS-2 or normal mouse IgG did not precipitate LAP activity in the sample. These results proved that the high LAP activity in the serum of the propositus is predominantly due to CD13.

isoelectric point of cd13
After isoelectric focusing, immunoblotting of the cellulose acetate membrane with anti-CD13 monoclonal antibody MCS-2 revealed a protein at pI 4.1 in the sera with normal LAP activities (from the mother, younger sister, and unrelated control). In contrast, MCS-2 revealed an additional band at pI 3.3 in sera with unusually high LAP activities from the propositus, her father, and her grandmother. Reactivity to this abnormal band was higher than that to the normal band at pI 4.1 in these samples (Fig. 2 A). Another anti-CD13 antibody, B-F10, also gave a similar result (data not shown). LAP activity could be detected only at pI 3.3. The enzyme activity at pI 4.1 in the sera with high LAP activity as well as normal samples was less than the sensitivity of the assay condition (Fig. 2B ). These results indicate that the abnormally high LAP activity is mainly derived from a variant CD13 with pI 3.3.

View larger version (46K): [in this window] [in a new window]   Figure 2. Immunoblotting with anti-CD13 (A) and LAP activity staining (B) of the cellulose acetate membranes on which high-voltage isoelectric focusing was performed. (A) After isoelectric focusing, one of the cellulose acetate membranes was reacted with anti-CD13 monoclonal antibody MCS-2. Serum samples of the propositus (lane 1), her father (lane 2), and paternal grandmother (lane 3) contained two distinct CD13 molecules, one with pI 3.3 and the other with pI 4.1, whereas her mother (lane 4), sister (lane 5), and a healthy control (lane 6) had a single CD13 with pI 4.1. (B) The other membrane was incubated with L-leucine-ß-naphthylamide. The abnormal CD13 molecule with pI 3.3 showed high LAP activity (lanes 1–3), whereas normal LAP activities were less than the sensitivity of this assay (lanes 4–6).

To characterize this variant CD13, we tested the effect of neuraminidase treatment of the sera on the isoelectric points. Interestingly, both the normal and variant CD13 showed pI 4.4 after treatment with neuraminidase (Fig. 3 A). This suggests that the variant CD13 contains more sialic acids than normal. Even after the pI shift, the variant CD13 still maintained high LAP activity (Fig. 3B ).

View larger version (29K): [in this window] [in a new window]   Figure 3. Effect of neuraminidase on isoelectric points. Partially purified serum CD13 from the propositus (P) and a healthy volunteer (N) were treated (lanes 3 and 4) or untreated (lanes 1 and 2) with neuraminidase. After high-voltage isoelectric focusing, the cellulose acetate membranes were reacted with MCS-2 (A) or incubated with L-leucine-ß-naphthylamide (B).

molecular mass of cd13
To determine the molecular mass of CD13, serum samples were subjected to immunoprecipitation followed by SDS-PAGE. In serum from a healthy volunteer, a 144 kDa-protein was precipitated by MCS-2 but not by normal mouse IgG (Fig. 4 A). Immunoprecipitation of serum from the propositus also showed a single band at the slightly larger position than normal, although the difference was subtle in this assay condition. After treatment with neuraminidase, the molecular mass of the normal CD13 diminished to 132 kDa, and the variant CD13 looked only slightly larger than normal (Fig. 4B ).

View larger version (49K): [in this window] [in a new window]   Figure 4. Immunoprecipitation of CD13 before and after treatment with neuraminidase. (A) Partially purified serum CD13 of the healthy control was immunoprecipitated with MCS-2 (lane 2) or normal mouse IgG and analyzed by SDS-PAGE. (B) Immunoprecipitated serum CD13 from the control (N) or the propositus (P) was treated (lanes 3 and 4) or untreated (lanes 1 and 2) with neuraminidase and analyzed by SDS-PAGE.

effects of n-glycosidase and o-glycosidase
Before the treatments with N-glycosidase and (or) O-glycosidase, a serum sample of the propositus and a partially purified normal CD13 sample were first treated with neuraminidase, immunoprecipitated, and further purified from the protein band excised from the SDS-PAGE gel. Being incubated with both N-glycosidase and O-glycosidase simultaneously, purified normal and the variant CD13 showed molecular masses of 100 kDa and 106 kDa, respectively (Fig. 5 ). Even treatment with N-glycosidase alone gave a similar result, whereas treatment with O-glycosidase alone had no effect on the molecular masses of CD13. These results suggest that the peptide of the variant CD13 is larger than normal and, in addition, virtually all carbohydrate chains of CD13 are covalently bound by N-glycosidic linkages.

View larger version (25K): [in this window] [in a new window]   Figure 5. Effect of glycosidases on molecular mass. The normal (N) and the variant serum CD13 (P) treated with neuraminidase were purified by immunoprecipitation and SDS-PAGE, then treated again with N-glycosidase and O-glycosidase simultaneously, and analyzed by SDS-PAGE.

	Discussion

Top Abstract Introduction Patient and Methods Results Discussion References

 
It might be interesting to test the cell surface LAP activity of the propositus because CD13 is a molecule originally defined as a marker of granulocyte–monocyte lineage, but the current study focused on analysis of the serum CD13 that showed extraordinarily high LAP activity by a conventional laboratory test.

The removal of 71% of LAP activity from the serum sample of the propositus by immunoprecipitation with an anti-CD13 monoclonal antibody and demonstration of the activity in the precipitate proved that high serum LAP activity of the propositus is predominantly due to CD13. This percentage is consistent with the report in which 60% to 80% of serum aminopeptidase activity was shown to be ascribed to CD13 in healthy individuals and patients with hepatobiliary disorders (6).

High-voltage isoelectric focusing followed by reaction with anti-CD13 monoclonal antibodies revealed that the isoelectric point of normal serum CD13 was 4.1. Only in the sera from family members with high LAP activity did we find a variant CD13, as a band with an abnormally low isoelectric point (pI 3.3), that was recognized with the two anti-CD13 monoclonal antibodies. The abnormal band showed extremely high enzyme activity by incubation with L-leucine-ß-naphthylamide, whereas the normal LAP activity that should be stained at pI 4.1 was much lower and hardly visualized. Thus the variant CD13 is considered to be the principal contributor to the unusually high serum LAP activity in this family case study. The variant CD13 in this case was inherited from the paternal grandmother of the propositus through the father to the propositus but was not passed on to her sister, and all members of the family, including the above members with the variant CD13, also had normal CD13 in their sera. Therefore the mode of inheritance is autosomal dominant and the members with high LAP activity in this family are heterozygotes of the variant and normal genes.

By treatment with neuraminidase, the isoelectric points of both normal and the variant CD13 became identical at pI 4.4, indicating that more sialic acids are bound to the variant CD13 molecule than to normal. The extra sialic acids of the variant CD13 may retard its turnover rate in serum, resulting in high serum LAP activity. In fact, a report compatible with this hypothesis has shown that desialylated serum glycoproteins have drastically reduced their survival times in the circulation (10).

From a search of the English literature, we found only one report of a family with high serum aminopeptidase activity (11)(12). However, in that case, the isoelectric point of the aminopeptidase of the propositus remained lower than normal even after treatment with neuraminidase, and it remained unclear whether the high serum aminopeptidase activity was derived from serum CD13 or from other enzymes.

CD13 is a glycoprotein with a molecular mass of 150 kDa, the peptide portion of which is 110 kDa (1). The deduced amino acid sequence of normal CD13 predicts 10 possible Asn-X-Ser/Thr N-glycosylation sites (3), and several O-glycosylation sites have also been suggested (13).

Glycoprotein molecular mass measurement by SDS-PAGE ought to be the most accurate when the carbohydrate chains are completely removed (14). In our results, the combined treatment with neuraminidase, N-glycosidase, and O-glycosidase decreased the molecular masses of CD13s, and the decreased size of the variant (106 kDa) was larger than normal (100 kDa). However, O-glycosidase alone caused no significant change in the size of CD13 treated with neuraminidase. These results suggest that the peptide of variant CD13 is longer than normal and virtually all of the carbohydrate chains are covalently bound to CD13 by N-glycosidic linkages. The extra portion of the peptide of the variant CD13 may contain extra glycosylation sites for carbohydrate chains, which would allow more addition of sialic acids to the variant CD13, explaining its lower isoelectric point.

The molecular masses of both normal (100 kDa) and the variant CD13 peptides (106 kDa) in serum seemed to be smaller than the reported molecular mass of CD13 peptide (110 kDa) directly extracted from cultured cells (1). Thus the peptides may be posttranslationally modified during or after the process of release from cell membrane into serum.

In conclusion, the high serum LAP activity in this family case is derived from a novel variant CD13 inherited in an autosomal dominant mode. The variant contains more sialic acids and is suggested to have a larger peptide than normal. Whether this variant is the product of a mutated CD13 gene or an abnormal modification after translation remains unknown. To elucidate the molecular mechanism that produces this variant, we are currently working to determine the cDNA sequence of this CD13 and the amino acid sequences of its N- and C-terminal portions.

	Footnotes

 
School of Allied Health Sciences and ¹ Department of Biochemistry, Faculty of Medicine, Tokyo Medical and Dental University, 5–45, Yushima 1-chome, Bunkyo-ku, Tokyo 113, Japan.

¹ Nonstandard abbreviations: LAP, leucine aminopeptidase; AAP, alanine aminopeptidase; PVDF, polyvinylidene difluoride; TBS, Tris-buffered saline; SDS-PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; and PIPES, piperazine-N,N'-bis-(2-ethanesulfonic acid).

	References

Top Abstract Introduction Patient and Methods Results Discussion References

 

1. Shipp MA, Look AT. Hematopoietic differentiation antigens that are membrane-associated enzymes: cutting is the key! Blood 1993;82:1052–70..
2. Olsen J, Cowell GM, Konigshofer E, Danielsen EM, Moller J, Laustsen L, et al. Complete amino acid sequence of human intestinal aminopeptidase N as deduced from cloned cDNA. FEBS Lett 1988;238:307-314. [ISI][Medline] [Order article via Infotrieve]
3. Look AT, Ashmun RA, Shapiro LH, Peiper SC. Human myeloid plasma membrane glycoprotein CD13 (gp150) is identical to aminopeptidase N. J Clin Invest 1989;83:1299-1307.
4. Razak K, Newland AC. The significance of aminopeptidases and haematopoietic cell differentiation. Blood Rev 1992;6:243-250. [ISI][Medline] [Order article via Infotrieve]
5. Gregg SL, Gajl-Peczalska KJ, LeBien TW, Bloomfield CD, Brunning R, Sagawa K. Monoclonal antibody MCS-2 as the marker of phorbol diester-induced myeloid differentiation in acute undifferentiated leukemia. Cancer Res 1984;44:2724-2730. [Abstract/Free Full Text]
6. Favaloro EJ, Browning T, Nandurkar H. The hepatobiliary disease marker serum alanine aminopeptidase predominantly comprises an isoform of the haematological myeloid differentiation antigen and leukaemia marker CD13/gp150. Clin Chim Acta 1993;220:81-90. [Medline] [Order article via Infotrieve]
7. Toda T, Sano-Shiba K, Cho H, Soon PL, Nakao M, Ohashi M. An improved method of high-voltage isoelectric focusing on cellulose acetate membranes. Electrophoresis 1988;9:149-150. [Medline] [Order article via Infotrieve]
8. Shiba K, Toda T, Iijima S, Inoue J, Yoshida T, Cho H, et al. Development of a thermoelectric cooling apparatus for high-voltage isoelectric focusing on a cellulose acetate membrane. Biol Pharm Bull 1994;17:1317-1320. [Medline] [Order article via Infotrieve]
9. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 1970;227:680-685. [Medline] [Order article via Infotrieve]
10. Ashwell G, Harford J. Carbohydrate-specific receptors of the liver. Ann Rev Biochem 1982;51:531-554. [ISI][Medline] [Order article via Infotrieve]
11. Hiwada K, Tokioka-Terao M, Nishimura K, Kokubu T. A family with a high serum aminopeptidase (microsomal) activity: properties of the enzyme from serum of the propositus. Clin Chem 1984;30:222-225. [Abstract]
12. Hiwada K, Ito T, Yokoyama M, Kokubu T. Isolation and characterization of membrane-bound arylamidases from human placenta and kidney. Eur J Biochem 1980;104:155-165. [Medline] [Order article via Infotrieve]
13. O'Connell PJ, Gerkis V, d'Apice AJF. Variable O-glycosylation of CD13 (aminopeptidase N). J Biol Chem 1991;266:4593-4597. [Abstract/Free Full Text]
14. Pitt-Rivers R, Impiombato FSA. The binding of sodium dodecyl sulfate to various proteins. Biochem J 1968;109:825-830. [ISI][Medline] [Order article via Infotrieve]