Biochemical Journal

Research article

Lamellipodin proline rich peptides associated with native plasma butyrylcholinesterase tetramers

He Li, Lawrence M. Schopfer, Patrick Masson, Oksana Lockridge

Abstract

BChE (butyrylcholinesterase) protects the cholinergic nervous system from organophosphorus nerve agents by scavenging these toxins. Recombinant human BChE produced from transgenic goat to treat nerve agent intoxication is currently under development. The therapeutic potential of BChE relies on its ability to stay in the circulation for a prolonged period, which in turn depends on maintaining tetrameric quaternary configuration. Native human plasma BChE consists of 98% tetramers and has a half-life (t½) of 11–14 days. BChE in the neuromuscular junctions and the central nervous system is anchored to membranes through interactions with ColQ (AChE-associated collagen tail protein) and PRiMA (proline-rich membrane anchor) proteins containing proline-rich domains. BChE prepared in cell culture is primarily monomeric, unless expressed in the presence of proline-rich peptides. We hypothesized that a poly-proline peptide is an intrinsic component of soluble plasma BChE tetramers, just as it is for membrane-bound BChE. We found that a series of proline-rich peptides was released from denatured human and horse plasma BChE. Eight peptides, with masses from 2072 to 2878 Da, were purified by HPLC and sequenced by electrospray ionization tandem MS and Edman degradation. All peptides derived from the same proline-rich core sequence PSPPLPPPPPPPPPPPPPPPPPPPPLP (mass 2663 Da) but varied in length at their N- and C-termini. The source of these peptides was identified through database searching as RAPH1 [Ras-associated and PH domains (pleckstrin homology domains)-containing protein 1; lamellipodin, gi:82581557]. A proline-rich peptide of 17 amino acids derived from lamellipodin drove the assembly of human BChE secreted from CHO (Chinese-hamster ovary) cells into tetramers. We propose that the proline-rich peptides organize the 4 subunits of BChE into a 340 kDa tetramer, by interacting with the C-terminal BChE tetramerization domain.

  • acetylcholinesterase (AChE)-associated collagen tail protein (ColQ)
  • butyrylcholinesterase (BChE)
  • lamellipodin
  • proline-rich attachment domain (PRAD)
  • proline-rich membrane anchor (PRiMA)
  • tetramer assembly

INTRODUCTION

BChE (butyrylcholinesterase; EC 3.1.1.8) is a serine hydrolase found in most vertebrate tissues and is especially abundant in serum, liver, intestine and lung [1,2]. BChE is capable of hydrolysing a wide range of choline and non-choline esters [3]. BChE has been suggested to have a role in controlling the activity of the neurotransmitter acetylcholine in brain as well as in the peripheral neuromuscular junction [2,46]. BChE protects the cholinergic nervous system from the toxicity of OP (organophosphorus toxin) nerve agents and pesticides by scavenging OP in a one-to-one stoichiometry. Purified human plasma BChE injected intravenously or intramuscularly into rodents and non-human primates has a prophylactic effect against acute and long-term toxicity of sarin, soman and VX {O-ethyl S-[2-(di-isopropylamino)ethyl]methylphosphonothioate} [79]. Recombinant human BChE produced in the milk of transgenic goats is a promising protein drug currently under development to treat nerve agent intoxication [10].

Approx. 98% of human BChE in serum is a tetramer of four identical subunits whose combined molecular mass is 340000 Da [3]. The molecule is a soluble globule, protected from proteolysis by a heavy sugar coating from nine N-linked carbohydrate chains. Human plasma BChE is synthesized in the liver and secreted into the blood where it has a half-life (t½) of 11–14 days [3,11]. The residence time of cholinesterases in blood shows a positive correlation with glycosylation state and tetrameric quaternary configuration, the latter factor being more crucial [12,13]. Wild-type human BChE expressed from HEK-293 cells (human embryonic kidney cells) and CHO (Chinese-hamster ovary) cell lines forms only approx. 10–30% tetramers. However, addition of poly(L-proline) to the culture medium or co-expression with the N-terminus of ColQ [AChE (acetylcholinesterase)-associated collagen tail protein] including the PRAD (proline-rich attachment domain) increased the amount of tetrameric BChE to 70% [14,15]. The PRAD-containing peptide co-purified with BChE tetramers [15], indicating that this proline-rich peptide not only drives the formation of tetramers but also is part of the final tetrameric complex. These observations raised the question of whether native plasma BChE tetramers contain a proline-rich peptide.

In the present study, we investigated the possibility that a poly-proline peptide is an intrinsic component of the native plasma BChE tetramer. Both purified human and horse BChEs were studied. The proteins were denatured by boiling or by lowering the pH with formic acid. Released peptides were purified by HPLC and analysed by MS. A series of peptides was identified by MALDI (matrix-assisted laser-desorption ionization)–TOF-MS (MALDI–time-of-flight MS). The same series of peptides was found in horse and human BChE samples. The peptides were sequenced by ESI-MS/MS (electrospray ionization tandem MS) and Edman degradation. These peptides all derived from a common, proline-rich sequence in lamellipodin (gi:82581557), which is a membrane-associated protein that participates in the regulation of lamellipodial protrusion, a component of directed cell motility [16]. When we added a proline-rich peptide of 17 amino acids derived from lamellipodin into the culture medium of CHO cells expressing wild-type human BChE, the peptide caused the formation of BChE tetramers.

EXPERIMENTAL

Human and horse BChEs

Human BChE (accession number gi:116353; Swiss-Prot database accession number P06276) was purified from plasma as described previously [17]. The purified human BChE was at least 90% pure. It was dissolved in 0.19 M NaCl, 20 mM Tris/HCl and 0.02% sodium azide (pH 7.5) to an activity of 1823 units/ml (2.5 mg/ml). Horse BChE was from Sigma (St. Louis, MO, U.S.A.). The purified horse BChE, freeze-dried out of PBS, was reconstituted by the addition of water to a concentration of 2.36 mg of BChE protein/ml with an activity of 1700 units/ml. Activity was assayed with 1 mM butyrylthiocholine in 0.1 M potassium phosphate buffer (pH 7.0) at 25 °C, in the presence of 0.5 mM DTNB [5,5′-dithiobis-(2-nitrobenzoic acid); Ellman's reagent]. One unit of activity is defined as micromoles of butyrylthiocholine hydrolysed per minute. Pure BChE has an activity of 720 units/mg.

Non-denaturing gradient gel electrophoresis

Polyacrylamide gradient gels (40–30%), 1.5 mm thick, were prepared in a Hoefer SE6000 gel apparatus (Hoefer Scientific Instruments, San Francisco, CA, U.S.A.). Electrophoresis was at 250 V constant voltage for 16 h at 4 °C. BChE samples were diluted with PBS (pH 7.2) and mixed with 50% (v/v) glycerol in 0.1 M Tris/HCl (pH 7.5) to a final glycerol concentration of 10%. The equivalent of 8 μg of human or horse BChE was loaded on to each lane for gel staining with Coomassie Brilliant Blue R-250 (Fisher Scientific).

Preparation of peptides for MALDI–TOF-MS

A 2 ml portion of 1823 units/ml human BChE and 2 ml of 1700 units/ml of horse BChE were put into dialysis bags with an MWCO (molecular-mass cut-off) of 3000 Da (Spectrum Laboratories, Rancho Dominguez, CA, U.S.A.) and dialysed against 4×4 litres of double-distilled water for 3 days at 4 °C. Approx. 2 ml of BChE solution was recovered from both human and horse samples after dialysis. Each BChE sample was divided into two fractions of equal volume. One fraction was heated in boiling water for 5 min to denature the protein. The horse BChE solution appeared turbid after heating, so it was briefly centrifuged and the precipitate was discarded. All samples were loaded into Centricon YM-10 centrifugal filters (MWCO=10 kDa; Millipore, Bedford, MA, U.S.A.) to separate peptides from protein. The centrifugation was in a Beckman CPR centrifuge with swinging bucket rotor at 3200 g for 40 min at 10 °C. The solution that passed through the filter membrane contained peptides and was recovered. Peptide solutions were concentrated to approx. 50–100 μl in a SpeedVac prior to MALDI analysis.

Reverse-phase HPLC

To prepare BChE peptide samples for sequencing and amino acid composition analysis, we used a Waters 625 HPLC system (Milford, MA, U.S.A.) equipped with a Zorbax 300 SB C-18 reverse-phase column (Agilent Technologies, Santa Clara, CA, U.S.A.). Purified human or horse BChE was boiled to release peptides. An alternative method for releasing peptides was to mix 1 mg of BChE in 0.5 ml with 0.5 ml of formic acid. Samples were filtered through a 0.2 μm syringe filter before injection into the HPLC.

The HPLC was operated at room temperature (22 °C) at a flow rate of 1 ml/min. Buffer A was 0.1% trifluoroacetic acid in water; buffer B was 0.07% trifluoroacetic acid in acetonitrile. Peptides were eluted with a gradient of 0–60% buffer B in 60 min. The absorbance was monitored at 220 nm. The HPLC eluent was collected into 1 min fractions and saved for analysis.

MALDI–TOF-MS

All MALDI–TOF-MS experiments were performed on an Applied Biosystems Voyager DE-PRO workstation equipped with a 337 nm pulsed nitrogen laser (Applied Biosystems, Framingham, MA, U.S.A.). A 1 ml portion of peptide sample was mixed 1:1 (v/v) with the matrix solution α-CHCA (α-cyano-4-hydroxycinnamate; 10 mg/ml in 50% (v/v) acetonitrile and 0.3% trifluoroacetic acid) on the MALDI target plate and allowed to dry at room temperature. Mass spectra were acquired in positive-ion, linear mode under delayed extraction conditions, using an acceleration voltage of 20 kV. Laser intensity was adjusted so that the strongest ion intensity in a spectrum did not exceed 80% of the maximum saturated intensity value. Laser positioning on the sample spot was monitored with a video camera. Spectra shown are the average of 500 laser shots collected from multiple locations on the target spot. Calibration for the mass spectra was performed externally using corticotropin peptides (amino acid residues 1–17, 18–39 and 7–38).

Q-Trap mass spectrometer

The amino acid sequence of peptides released from human and horse BChEs was determined by collision-activated dissociation in a Q-Trap 2000, a hybrid quadrupole linear ion trap mass spectrometer equipped with a nanospray interface (Applied Biosystems). The spectrometer was calibrated daily on selected fragments from the collision-activated spectrum of [glutamic acid]fibrinopeptide B. HPLC eluent fractions containing target peptides (identified by MALDI–TOF-MS) were reconstituted into 60% (v/v) acetonitrile and 0.1% formic acid and then loaded into a silver-coated nano-infusion emitter (Econo12; New Objective, Woburn, MA, U.S.A.) using a gel-loading pipette tip (GELoader, Eppendorf, Westbury, NY, U.S.A.). The emitter can hold up to 12 μl of sample. The loaded emitter was fitted into the Discrete Nanospray™ head of the nanospray source on the Q-Trap mass spectrometer. Peptides were introduced into the mass spectrometer by means of static infusion, at room temperature, using an ion spray voltage of 1300 V (which creates a voltage differential of 1300 V between the emitter and the curtain plate). The emitter position was optimized to obtain maximum signal intensity. Mass spectra were collected in the enhanced mode, i.e. using the ion trap. Enhanced mass spectra were collected to identify the target peptide. EPI (enhanced product ion) was used for fragmenting the target peptide. The collision cell was pressurized to 40 μtorr (1 torr=0.133 kPa) with pure nitrogen and collision energies of 40–50 V were used. The default trap fill time for each EPI scan was 20 ms. A total of 200 EPI scans were accumulated to generate the final EPI spectrum. The EPI spectra of the target peptide were manually analysed to determine the sequence of the peptide.

Amino acid composition analysis

Peptide samples were subjected to acid hydrolysis in which peptides were heated at 110 °C in 6 M HCl for 20 h. Composition analysis was performed by a Hitachi L-8800 amino acid analyser (Hitachi, Tokyo, Japan), using post-column derivatization, as previously described [18].

Edman degradation sequencing of peptides

A Procise 494 N-terminal sequencer (Applied Biosystems) was used to sequence peptides by Edman degradation. Standard company settings and running cycles were applied. Samples were immobilized on to a PVDF membrane for the experiment. A detailed description of the method can be found in [19].

Lamellipodin proline-rich peptide experiment

A 17-amino-acid proline-rich peptide derived from lamellipodin was custom-synthesized by GenScript (Piscataway, NJ, U.S.A.). The sequence of the peptide was: PPPPPPPPPPPPPPPLP. The peptide, dissolved in water and filter-sterilized, was added to the culture medium of CHO cells to final concentrations of 5 and 50 μM. The CHO cells used in this experiment express and secrete wild-type human BChE as described previously [15]. A 20 μl portion of culture medium (with or without the proline-rich peptide), along with 20 μl of human plasma as a control, were loaded on to a non-denaturing polyacrylamide gel and electrophoresed at 250 V constant voltage for 16 h at 4 °C. The gel was stained for BChE activity by using the method of Karnovsky and Roots [20].

RESULTS

Peptides were released from denatured BChE tetramers

The purity of the BChE preparations used in the present study was determined by non-denaturing PAGE. Coomassie Brilliant Blue staining of the gel showed a heavy band migrating at the position of tetrameric BChE (for the gel picture, see Supplementary material at http://www.BiochemJ.org/bj/411/bj4110425add.htm). No other bands can be clearly seen on the gel. The absence of an albumin band from both samples rules out the possibility that the peptides released by boiling originated from albumin. This was a concern because albumin sequesters low-molecular-mass peptide fragments in blood [21].

Peptides released by boiling were separated from BChE protein on a centrifugal filter with a 10000 Da cut-off. The pass-through solution was collected and analysed by MALDI–TOF-MS. Control native BChE samples, treated identically except that they were not boiled, were also analysed by MALDI–TOF. As shown in Figure 1(A), the pass-through solution from native human BChE sample (upper panel) gave virtually no peptides, whereas the pass-through solution from boiled human BChE (lower panel) gave a series of peptides, ranging from approx. 2000 Da to 3000 Da. Similar observations were made for the horse BChE samples (Figure 1B).

Figure 1 MALDI mass spectra of peptides released from native or denatured plasma BChE

Purified plasma BChE from either human (A) or horse (B) was divided into two fractions. One fraction was boiled and the other was not boiled. Peptides were separated from the native and denatured proteins and subjected to MALDI analysis. Native human and horse BChE samples (upper panels of A and B) released no peptides. A series of peptides can be identified from both denatured human and horse BChE samples (lower panels of A and B).

Table 1 lists the masses of the peptides released from horse and human BChEs observed in the MALDI spectra. The differences between masses show that most of these peptides differ by single amino acids. For example, in the human preparation, the difference between the 2171.5 and 2074.3 masses is 97.2 a.m.u. (atomic mass unit), which is the dehydro-mass of proline [19]. The difference between the 2171.5 and 2285.1 masses is 113.6 (leucine or isoleucine), between 2452.8 and 2566.7 is 113.9 (asparagine), between 2566.7 and 2663.7 is 97.0 (proline) and between 2663.7 and 2878.5 is 214.8 (consistent with serine plus glutamine). These observations suggest that these masses represent a family of related peptides and not a collection of wholly different sequences. Similar observations can be made for the peptides from the horse BChE preparation.

View this table:
Table 1 Summary of the peptide masses released from denatured human and horse plasma BChE tetramers

Only peaks greater than 2000 Da are recorded in the Table. Each peak is the most intense peak in a group of three related peaks. A detailed explanation for the nature of the peaks in each group is given in the text.

The denatured samples from both horse and human BChEs contain nearly identical sets of peptides (Table 1), even though they are from two different species and two different preparations. These results strongly suggest that the peptides are not due to contamination of the BChE samples during purification. The fact that this series of peptides can only be found after denaturation of BChE further suggests that the peptides are part of the BChE tetrameric structure and are not adventitiously bound to the surface of BChE. Finally, the peptides are not attached through disulfide bonds, as no reducing agent was required for their release.

One final point concerning the complexity of the MALDI spectrum should be mentioned. Each labelled peptide in Figure 2 is the most intense peak in a group of three peaks. The second most intense peak is 22 a.m.u. heavier than the first and the least intense peak is 22 a.m.u. heavier than the second. This is consistent with sodium ion adducts contributing the positive charge, instead of a hydrogen ion.

Figure 2 Sequencing of peptide 2566 (A) and peptide 2663 (B) by using ESI-MS/MS fragment ion spectra

The peak labelled with an asterisk in (A) has an m/z value of 1283.6, which represents the doubly charged state of peptide 2566. The peak labelled with an asterisk in (B) has an m/z value of 1332.3, which represents the doubly charged state of peptide 2663. Major y-series fragment ions used in sequencing the peptides are labelled. Major b-series product ions can also be readily identified in the two spectra and were used to sequence the peptides from the N-terminus. cps, counts per s.

The peptides released from denatured BChE are proline-rich

Peptides released from BChE by boiling were separated from BChE protein and partially resolved from one another by HPLC. HPLC-purified peptides were electrosprayed into the Q-Trap mass spectrometer where fragment ion spectra of selected peptides were generated through collision-activated dissociation. Figures 2(A) and 2(B) show the fragment ion spectra of peptides 2566 and 2663 respectively. The peptide sequences were determined from the fragment ion spectra.

The spectrum of peptide 2566 revealed y-series ions that contained 12 proline residues in a row with a Pro(Leu/Ile)Pro sequence at the C-terminus. The b-ion series, although present in the spectrum, was less easily interpreted. This is because several different b-series sequences of equivalent intensity could be identified. We attribute this multitude of sequences to internal fragmentation. Each of these sequences included an N-terminal residue (either proline or serine), but the true N-terminus could not be unequivocally assigned. A leucine/isoleucine was also frequently observed in the first five N-terminal residues. After the first few N-terminal residues, each b-ion sequence became a string of proline residues. Up to 12 proline residues in a row were observed. Because of this complexity, the complete sequence of this peptide could not be determined from the Q-Trap data alone. However, the sequence data together with the peptide mass suggested that peptide 2566 contained at least 20 consecutive proline residues, as well as one serine and two leucine/isoleucine residues. The fragmentation pattern of peptide 2663 was nearly identical with that of peptide 2566. It also revealed a Pro(Leu/Ile)Pro C-terminus followed by a string of proline residues (at least 19). The similarity of these fragmentation patterns supports the proposal that these peptides are representative members of a family.

The HPLC fraction containing peptides 2566 and 2663 was further purified by using reverse-phase HPLC. The resulting sample, which contained only peptides 2566 and 2663 (Figure 3), was subjected to amino acid composition analysis and amino acid sequencing by Edman degradation. The composition analysis data indicated that the unit composition of the sample was 24 proline, two leucine and one serine residues. This is comparable with the composition obtained from the mass spectral sequence analysis.

Figure 3 MALDI mass spectrum of human BChE-associated peptides

This peptide sample was purified on a C-18 reverse-phase HPLC column (see the Experimental section) and sequenced by Edman degradation.

Edman degradation generated the complete sequence of the two peptides: the sequence of one peptide was SPPLPPPPPPPPPPPPPPPPPPPPLP, and the sequence of the other peptide was PSPPLPPPPPPPPPPPPPPPPPPPPLP. The two sequences match the sequences for the peptides 2566 and 2663 obtained by MS and confirm that they indeed differ only by an N-terminal proline residue.

Similar purification and sequencing experiments were applied to the horse BChE peptide samples (results not shown) and the results showed that horse peptides 2566 and 2663 had exactly the same sequences as their human counterparts.

ESI-MS/MS sequencing was also performed on other peptides in Table 1. All peptides contained a string of proline residues, indicating the poly-proline core sequence. This supports the proposal that they all derived from a common source protein.

Peptides released from denatured BChE originate from RAPH1 [Ras-associated and PH domain (pleckstrin homology domain)-containing protein 1], also called lamellipodin

Peptide PSPPLPPPPPPPPPPPPPPPPPPPPLP (molecular mass=2663.2 Da) was used to query the NCBI (National Center for Biotechnology Information) non-redundant human protein database by using BLAST. RAPH1 (gi:82581557) was the only protein that exactly matched the query sequence. This protein is also called lamellipodin, or proline-rich EVH1 [Ena/VASP (vasodilator-stimulated phosphoprotein) homology 1] ligand 2 [16]. The partial sequence information that we obtained from all the other peptides also fit the lamellipodin sequence. Lamellipodin is thus identified as the source of the peptides released from denatured human BChE.

By combining the peptide masses from the MALDI–TOF-MS with the ESI-MS/MS sequencing results and the sequence of human lamellipodin, we were able to obtain complete sequences that were consistent with all of the masses observed in the MALDI spectra from the denatured BChE tetramers, except one. Only the sequence of peptide 3158 was not identified. No portion of the proline-rich core sequence of lamellipodin yielded this mass. A list of the peptides is given in Table 2.

View this table:
Table 2 Amino acid sequences of BChE-associated peptides

There is no horse protein database available for a BLAST-like query at this time. However, given the observed similarity between the peptide masses and peptide sequences from human and horse BChEs, it is reasonable to predict that peptides released from horse BChE originate from a horse protein closely related to human lamellipodin.

Proline-rich peptide derived from lamellipodin caused assembly of BChE into tetramers

Previous studies showed that addition of poly(L-proline) peptide into the culture media of CHO cells expressing wild-type human BChE increased the proportion of BChE tetramers and decreased the amount of dimers and monomers [15]. To investigate whether the proline-rich peptides we identified from plasma BChE can drive the formation of BChE tetramers, we expressed wild-type human BChE in CHO cells in the presence of a 17-mer proline-rich peptide with the sequence PPPPPPPPPPPPPPPLP. This peptide accounts for amino acids 686–702 in the human lamellipodin sequence (gi:82581557). As shown in Figure 4, addition of the lamellipodin-derived proline-rich peptide to the culture medium increased the proportion of BChE tetramers from approx. 10% to approx. 60% (5 μM final peptide concentration in the culture medium) or 80% (50 μM final peptide concentration in the culture medium). The proportion of BChE dimers and monomers in the culture medium was correspondingly decreased. These results indicated that the proline-rich peptides we found in plasma BChE can guide the tetramerization of BChE.

Figure 4 Lamellipodin proline-rich peptide promotes BChE tetramerization

A non-denaturing gel was stained for BChE activity. Lane 1, human plasma; lane 2, culture medium without proline-rich peptide; lane 3, culture medium with 5 μM proline-rich peptide; lane 4, culture medium with 50 μM proline-rich peptide.

DISCUSSION

The tetramer-organizing function of proline-rich peptides

Our finding that the serum BChE tetramer is associated with a family of proline-rich peptides derived from lamellipodin establishes the identity of the tetramer-organizing peptide as PSPPLPPPPPPPPPPPPPPPPPPPPLP (and variations thereof). It thereby defines a new, naturally occurring tetramerization peptide. Previously, membrane-bound forms of BChE and AChE have been shown to be associated with the proline-rich regions of ColQ and PRiMA (proline-rich membrane anchor) [22,23]. Our finding reinforces the requirement for a proline-rich peptide in the formation of tetrameric cholinesterase structures. Also, it opens the possibility that other, yet unidentified forms of tetramer-organizing peptide may exist for cholinesterases in other tissues.

Association of BChE with lamellipodin-derived peptide

Plasma BChE is secreted into the blood from the liver. Secreted proteins are synthesized in the rough ER (endoplasmic reticulum) and transported into the Golgi complex where they are glycosylated, assembled and packed into vesicles that are destined for secretion. Lamellipodin, on the other hand, is a cytosolic protein [16]. The molecular pathway that directs the association of BChE and lamellipodin-derived proline-rich peptide is unclear. However, the events leading to the generation of peptides used by MHC class I molecules for cellular immune response may provide a hint for the answer.

Approximately one-third of newly synthesized proteins are rapidly degraded by proteasomes in the cytosol under physiological conditions [24]. The peptides needed for antigen presentation are generated in the cytosol by cleavage of endogenous proteins and are displayed on the cell surface by MHC class I molecules, which are present on nearly all nucleated cells [25]. TAPs (transporters associated with antigen processing) are located on the ER membrane and are responsible for moving peptides from the cytosol into the ER for binding with MHC class I molecules. TAPs translocate peptides with broad specificity. Although it is most efficient for peptides of 8–12 amino acids, peptides of up to 40 amino acids can also be transported into the ER by TAPs [26]. It is conceivable that the cytosolic protein lamellipodin is degraded in the cytosol by proteasomes and the resulting peptides, including the proline-rich peptides we identified in the present study, are transported into ER by TAPs, where they associate with newly synthesized BChE molecules to form tetramers.

The heterogeneity of the proline-rich peptide

A total of eight proline-rich peptides were identified from human and horse BChE samples, all inter-related and originating from the same protein (see Table 2). The proteasome in the cytosol is a protein complex and can cleave at nearly every position of a protein [27]. Other proteases in the cytosol, such as IFN (interferon)-γ-inducible leucine aminopeptidase and tripeptidyl peptidase II, have been shown to work along with proteasomes in processing peptides to be transported into the ER by TAPs [28,29]. All the above-mentioned proteases may contribute to produce a mixture of lamellipodin-derived proline-rich peptides that are imported into the ER for interaction with BChE.

Alternatively, the peptides we observed might all derive from a single progenitor peptide that was progressively proteolysed, from both the N- and C-termini, while attached to the BChE tetramer during circulation in the blood. Koomen et al. [30] found progressive N- and C-terminal proteolysis of naturally occurring peptides by aminopeptidases and carboxypeptidases in blood to be the general rule. This resulted in families of peptides with variously truncated N- and C-termini [30]. However, if the observed heterogeneity in the proline-rich peptides is due to proteolysis in the circulation, then that proteolysis would have had to have happened while the peptide was bound to BChE. Dvir et al. [22] showed that the 14-residue ColQ PRAD peptide (LLTPPPPPLFPPPFF) was longer than the tetramerization domain of AChE and the length of the tetramerization domain was about the same as the thickness of the AChE tetramer. Furthermore, they showed that if a larger proline-rich peptide was complexed with AChE, the C-terminus could pass easily between the AChE subunits in one direction while the N-terminus could pass out beyond the end of the tetramerization domain in the other direction. BChE is nearly identical in structure with AChE [31]. Therefore the AChE models should apply equally well to BChE. We found various proline-rich peptides attached to BChE that included residues 676–714 from lamellipodin. That suggests that the original peptide may have been up to 39 residues long. As such, it should have been long enough to extend beyond the protection of the BChE molecule in both directions, making it accessible to the peptidases in blood, and thus susceptible to degradation during circulation.

C5 genetic variant of human BChE

Approx. 8% of Caucasians have a variant form of serum BChE, named C5, which migrates more slowly than the normal serum BChE during electrophoresis on non-denaturing polyacrylamide gels [32,33]. Studies on the C5 BChE showed that it consists of the normal BChE tetramer non-covalently associated with an unknown protein component with an estimated molecular mass of 60 kDa. The coding gene for this unknown protein is located on chromosome 2q33–q35 [34]. The human lamellipodin gene is also located on chromosome 2q33 (obtained from NCBI GENE database under the gene name RAPH1). This striking coincidence suggests that the unknown protein component of C5 BChE could actually be a truncated form of lamellipodin. If this is true, the presence of C5 BChE in blood may suggest malfunctioning of a certain protease or a protein-processing pathway in the liver. Although C5 BChE carriers do not show any pathology, further investigation in this direction may have broader clinical significance.

Tetrameric BChE has a longer half-life in the circulation

Proteins and peptides in blood are constantly subjected to renal clearance and enzymatic degradation. Their circulation half-life ranges from a few minutes to several days [35]. The therapeutic potential of BChE against organophosphate toxicants (nerve agents and pesticides) relies heavily on its ability to stay in the circulation for a prolonged period. Native and recombinant BChE tetramers showed a half-life of 16–56 h in rodents in various studies, whereas recombinant human BChE monomers and dimers only have a half-life of 2–300 min [12,36]. Therefore expression of the tetrameric form of recombinant BChE is an important factor in creating a viable therapeutic agent. Our discovery of the native proline-rich peptide that is responsible for the formation of the human serum BChE tetramer should contribute to the creation of an expressed, tetrameric form of BChE that will be stable in the circulation. This fulfils critical criteria in developing BChE into an antinerve agent therapeutic.

Acknowledgments

MS, Edman degradation and amino acid composition analysis were performed with the support of the Protein Structure Core Facility at the University of Nebraska Medical Center. This work was supported by U.S. Army Medical Research and Materiel Command contract W81XWH-06-1-0102, Edgewood Biological Chemical Center contract W911SR-04-C-0019, Eppley Cancer Center grant P30CA36727 and NIH (National Institutes of Health) grant 1 U01 NS058056-01.

Abbreviations: AChE, acetylcholinesterase; a.m.u., atomic mass unit; BChE, butyrylcholinesterase; CHO, Chinese-hamster ovary; ColQ, AChE-associated collagen tail protein; EPI, enhanced product ion; ER, endoplasmic reticulum; ESI-MS/MS, electrospray ionization tandem MS; OP, organophosphorus toxin; PRAD, proline-rich attachment domain; PRiMA, proline-rich membrane anchor; MALDI, matrix-assisted laser-desorption ionization; MALDI–TOF-MS, MALDI–time-of-flight MS; MWCO, molecular-mass cut-off; PH domain, pleckstrin homology domain; RAPH1, Ras-associated and PH domain-containing protein 1; TAP, transporter associated with antigen processing

References

View Abstract