Báo cáo khoa học: Structure and inﬂuence on stability and activity of the N-terminal propeptide part of lung surfactant protein C

Structure and inﬂuence on stability and activity of the N-terminal propeptide part of lung surfactant protein C Jing Li1, Edvards Liepinsh1, Andreas Almle´ n2, Johan Thyberg3, Tore Curstedt2, Hans Jo¨ rnvall1 and Jan Johansson4

1 Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden 2 Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden 3 Department of Cell and Molecular Biology, Karolinska Institutet, Stockholm, Sweden 4 Department of Molecular Biosciences, Swedish University of Agricultural Sciences, The Biomedical Centre, Uppsala, Sweden

Keywords surfactant protein; peptide aggregation; proprotein processing; peptide structure analysis; surface activity

Correspondence J. Li, Department of Medical Biochemistry and Biophysics, Karolinska Institutet, S-17177 Stockholm, Sweden Fax: +46 8337462 Tel: +46 852487681 E-mail: jing.li@mbb.ki.se

Note The coordinates for the NMR structure of SP-Ci(1–31) in DPC micelle have been deposited in the protein data bank under the accession code 2esy.

independent of pH,

(Received 15 November 2005, revised 20 December 2005, accepted 20 December 2005)

doi:10.1111/j.1742-4658.2006.05124.x

Mature lung surfactant protein C (SP-C) corresponds to residues 24–58 of the 21 kDa proSP-C. A late processing intermediate, SP-Ci, corresponding to residues 12–58 of proSP-C, lacks the surface activity of SP-C, and the SP-Ci a-helical structure does not unfold in contrast to the metastable nature of the SP-C helix. The NMR structure of an analogue of SP-Ci, SP-Ci(1–31), with two palmitoylCys replaced by Phe and four Val replaced by Leu, in dodecylphosphocholine micelles and in ethanol shows that its a-helix vs. that of SP-C is extended N-terminally. The Arg-Phe part in SP-Ci that is cleaved to generate SP-C is localized in a turn structure, which is followed by a short segment in extended conformation. Circular dichroism spectroscopy of SP-Ci(1–31) in microsomal or surfactant lipids shows a mixture of helical and extended conformation at pH 6, and a shift to more unordered structure at pH 5. Replacement of the N-terminal hexa- peptide segment SPPDYS (known to constitute a signal in intracellular targeting) of SP-Ci with AAAAAA results in a peptide that is mainly unstructured, in microsomal and surfactant lipids. Addition of a synthetic dodecapeptide, corresponding to the propeptide part of SP-Ci, to mature SP-C results in slower aggregation kinetics and altered amyloid ﬁbril formation, and reduces the surface activity of phos- pholipid-bound SP-C. These data suggest that the propeptide part of SP-Ci prevents unfolding by locking the N-terminal part of the helix, and that acidic pH results in structural disordering of the region that is proteolyti- cally cleaved to generate SP-C.

accepted that the hydrophobic SP-B and SP-C are pri- marily involved in the reduction of alveolar surface tension [1–3], while the hydrophilic SP-A and SP-D are involved in pulmonary innate host defence [4–6].

Pulmonary surfactant is a lipid ⁄ protein mixture that reduces surface tension and exerts host defence func- tions at the alveolar air ⁄ liquid interface. Surfactant contains mainly phospholipids but also cholesterol and small amounts of fatty acids. Four surfactant-associ- ated proteins, surfactant protein A (SP-A), SP-B, SP-C and SP-D, have been described [1,2]. It is generally

SP-C is a 35-residue transmembrane lipopeptide uniquely expressed in the alveolar type II cell [7]. The function of SP-C in vivo remains unclear, but effects of

Abbreviations DPC, [2H38]dodecylphosphocholine; DPPC, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine; ER, endoplasmic reticulum; LPS, lipopolysaccharide; POPG, 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphoglycerol; SP, surfactant protein; SP-Ci, surfactant protein C processing intermediate; TFA, triﬂuoroacetic acid.

FEBS Journal 273 (2006) 926–935 ª 2006 The Authors Journal compilation ª 2006 FEBS

926

J. Li et al.

Properties of SP-C N-terminal propeptide

both surface activity and LPS binding. This, in combi- nation with the almost complete absence of mature SP-C, suggests that SP-B-deﬁcient children lack active forms of both SP-B and SP-C [28].

species

structures of different

The presence of the dodecapeptide propeptide part in SP-Ci also prevents the a-helix from unfolding in neutral aqueous organic solvents for at least 1 month, while in acidiﬁed solvents SP-Ci aggregates and forms amyloid-like ﬁbrils in a few days [29]. This suggests that pH affects the conformation of SP-Ci, which is intriguing as the pH is neutral in the early secretory pathway, while the pH of lamellar bodies is 5.5–6 [30]. In the present work we further investigated the struc- ture of SP-Ci and how the dodecapeptide N-terminal ﬂanking peptide affects SP-C stability and activity.

Results

NMR structure of SP-Ci(1–31) in DPC micelles

SP-C on the adsorption, spreading, and stability of lipid ﬁlms at an air ⁄ water interface have been docu- mented in a number of in vitro studies. In addition, it has been shown recently that SP-C can recognize lipo- polysaccharide (LPS) [8]. The SP-C primary and secon- dary are highly conserved. SP-C is composed of a ﬂexible N-terminal end, where two palmitoyl groups are linked to cysteine residues at positions 5 and 6, and an a-helical C-ter- minal part [9,10]. The SP-C a-helix is metastable and can irreversibly transform into b-sheet aggregates, forming amyloid ﬁbrils under pathological conditions, or during incubation in aqueous organic solvents the SP-C palmitoyl groups [11–15]. Removal of accelerates aggregation [16]. However, the detailed mechanisms underlying SP-C ﬁbril formation are not clear. The SP-C helix contains a polyvaline segment (containing 10–12 valine residues out of 16) which is intriguing since valine has the highest b-strand propen- sity of all residues. This feature likely underlies the tendency of SP-C to unfold and aggregate, since replacement of valine with leucine (which favours helix formation) results in a stable a-helix [17,18].

Due to the rare occurrence of SP-B deﬁciency, it is dif- ﬁcult to obtain sufﬁcient amounts of SP-Ci for NMR analysis. Therefore, SP-Ci(1–31) (Table 1) was synthes- ized for structural studies. Sequence-speciﬁc 1H-NMR assignments of SP-Ci(1–31) incorporated in [2H38]dode- cylphosphocholine (DPC) micelles were obtained using a series of NOESY, TOCSY and COSY spectra. All spin systems could be assigned and the chemical shifts are given in supplementary Table S1. 1H-NMR assign- ments of SP-Ci(1–31) dissolved in ethanol were also obtained (supplementary Table S2). The chemical shifts of SP-Ci(1–31) in the two solvent systems are similar, the Ha shifts differ by <0.08 p.p.m.

compartments

the maximum NOE restraint

Mature SP-C corresponds to residues 24–58 of the 197-residue integral membrane protein proSP-C, and is generated via multiple proteolytic cleavages [19]. The ﬁnal cleavage of proSP-C to SP-C likely occurs in the lysosome-like lamellar bodies in which surfactant lipids and proteins are packed together, and from which they are secreted by exocytosis into the alveolar space [20,21]. ProSP-C has a type II orientation in the endo- plasmic reticulum (ER), exposing the N-terminal part to the cytosol and the C-terminal part to the ER lumen. The correct intracellular trafﬁcking of proSP-C has been reported to depend on the N-terminal propeptide. Deletion of the region from residues Met10 to Thr18 of proSP-C results in retention of the truncation peptide in the ER [21,22]. In contrast, mutants of proSP-C, which lack the C-terminal propart, are directed to distal in transfected epithelial cells [22].

For deﬁnition of the structure of SP-Ci(1–31) in DPC micelles, a set of 263 nonredundant NOE-derived upper distance limits and 81 dihedral angle restraints were used for structure calculations using the program dyana, followed by energy minimization in vacuo with the program opal. For the 20 best conformers after minimization the amber energy was )915 ± 25 kcalÆ mol)1, violations 0.10 ± 0 A˚ , the maximum dihedral angle restraint vio- lations 2.05 ± 0.18 degrees, the rmsd to the mean for

Table 1. Amino acid sequences of SP-C, SP-Ci and analogues. The Cys in SP-C and SP-Ci are palmitoylated.

SP-C SP-Ci

SP-Ci(1–31) AlaSP-Ci(1–31) SP-C33:

FGIPCCPVHLKRLLIVVVVVVLIVVVIVGALLMGL SPPDYSAAPRGRFGIPCCPVHLKRLLIVVVVVVLI VVVIVGALLMGL SPPDYSAAPRGRFGIPFFPVHLKRLLILLLL AAAAAAAAPRGRFGIPFFPVHLKRLLILLLL IPSSPVHLKRLKLLLLLLLLILLLILGALLMGL

Infants with inherited deﬁciency of SP-B show neo- natal respiratory distress, which is refractory to treat- ment with exogenous surfactant, and the only effective treatment is lung transplantation [23–25]. In addition, poorly formed lamellar bodies and accumulation of an SP-C precursor are observed in SP-B deﬁciency [26,27]. This processing intermediate, SP-Ci, shares immunore- activity with SP-C and was recently found to contain an N-terminal dodecapeptide extension (corresponding to proSP-C residues 12–23), but is otherwise identical to mature SP-C [28]. The presence of the N-terminal dodecapeptide in SP-Ci results in strong inhibition of

FEBS Journal 273 (2006) 926–935 ª 2006 The Authors Journal compilation ª 2006 FEBS

927

J. Li et al.

Properties of SP-C N-terminal propeptide

mature SP-C) as compared to SP-C(1–17) are in agree- ment with an N-terminal extension of the a-helical structure [32].

CD studies of SP-Ci(1–31) and AlaSP-Ci(1–31) at different pH

Fig. 1. Stereoview of the backbone heavy atoms for the 20 best conformers obtained from the structure calculation of SP-Ci(1–31) in DPC micelles.

shown). Similar

The stability of SP-Ci is pH dependent, which promp- ted us to study the structure of SP-Ci(1–31) at different pH. At pH 6, SP-Ci(1–31) in microsome lipids shows a CD spectrum that indicates the presence of both a-helix and b-strand structures (Fig. 3). An increase in random structure, as judged from the shift to lower wavelengths of the broad minimum was observed when the pH was lowered to 5 (Fig. 3). Spectra recorded at pH 4 and 3 were similar to the pH 5 spectrum (data results were obtained with not SP-Ci(1–31) incorporated in surfactant lipids (data not shown). Due to solubility problems, CD spectra recor- ded at pH 7 gave low signal : noise ratios.

AlaSP-Ci(1–31) in microsome lipids at pH 6 gives a CD spectrum which indicates the presence of random and helical structures (Fig. 3), and no change is seen when the pH is lowered to 5. As for SP-Ci(1–31), very similar results were obtained when the peptide was incorporated in surfactant lipids.

Aggregation and ﬁbril formation of SP-C with and without the propeptide

backbone heavy atoms in the region encompassing res- idues 12–31 0.70 ± 0.35 A˚ , and the rmsd for all heavy atoms in the region 12–31 1.55 ± 0.71 A˚ . Figure 1 shows the heavy atoms for the superimposed 20 best conformers. The structure of SP-Ci(1–31) in DPC micelles is disordered from Ser1 to Gly11, shows a turn in the region Arg12 to Phe13, an extended stretch involving Gly14 and Ile15, a turn from Pro16 to Phe18, and an a-helix from Pro19 to Leu31. The a-helix of SP-Ci(1–31) in DPC micelles thus starts four residues N-terminally of the helix of a synthetic pep- tide corresponding to residues 1–17 of mature SP-C in DPC micelles, the latter starting at Lys11 [31], corres- ponding to Lys23 of SP-Ci(1–31). This conclusion was conﬁrmed by chemical shift analysis for Ha protons in comparable parts of the two peptides (Fig. 2). Signiﬁ- cant high ﬁeld shifts for residues Pro19 and Val20 of SP-Ci(1–31) (corresponding to Pro7 and Val8 of

0.3

0.2

The more stable a-helix and lower surface activity of SP-Ci compared to SP-C may be a result of noncova- lent interactions between the N-terminal dodecapeptide part and the SP-Ci helix. A ﬁvefold molar excess of a synthetic peptide corresponding to the N-terminal dodecapeptide of SP-Ci was incubated with SP-C in aqueous organic solvents. Peptide aggregation was

]

0.1

m p p

40 ) 1 - l

y t i c i t p

20 SP-Ci(1-31), pH=6 SP-Ci(1-31), pH=5 AlaSP-Ci(1-31), pH6

-0.1

[ ) α H ( ω ∆

0 o m d x 2

-0.2

i l l e r a l o m

-0.3

-20

F G I P C C P V H L K R L L

-40

m c x g e d k (

l a u d i s e R

200

210

220

230

240

250

260 Wavelength (nm)

Fig. 2. Ha chemical shifts of SP-Ci(1–31)–Ha shifts of SP-C(1–17) in DPC micelles. The shift differences are shown for residues 1–14 of SP-C, and where the two studied peptides have identical residues. Nonidentical residues in the two peptides are due to species differ- ences [porcine SP-C(1–17) and human SP-Ci(1–31)]. In the latter peptide Cys at positions 5 and 6 are replaced with Phe.

Fig. 3. CD spectra of SP-Ci(1–31) and AlaSP-Ci(1–31) in microsomal lipids at different pH.

FEBS Journal 273 (2006) 926–935 ª 2006 The Authors Journal compilation ª 2006 FEBS

928

J. Li et al.

Properties of SP-C N-terminal propeptide

experiment, and a decrease

recording MALDI MS over time, and normalizing the ion intensities to a nonaggregating internal standard, SP-C33. SP-C incubated with the dodecapeptide shows slower aggregation kinetics than SP-C alone (Fig. 4).

Electron micrographs of

determined via measurements of the rate of disappear- ance of soluble protein, as described for SP-C [18,29]. The insoluble protein species are not ionized in a MALDI in protein [M+H]+ ion current will be related to a decrease of protein concentration in solution and an increase in insoluble matter, i.e. aggregated protein. The stability of SP-C in 95% aqueous ethanol was analysed by

insoluble material were taken after 168 h incubation (Fig. 5). For SP-C alone, typical amyloid-like ﬁbrils were visible. However, SP-C incubated with the dodecapeptide did not form ﬁbrils.

SP-C + dodecapeptide

SP-C alone

A

100

SP-C

SP-C33

T=0 h

SP-C33

y t i s n e t n

y t i s n e t n I

3500

4500

5000

0 3000

3500

4500

5000

0 3000

4000 Mass (m/z)

100

SP-C

SP-C33

SP-C

T=120 h

SP-C33

y t i s n e t n I

0 3000

3500

4500

5000

3500

4500

5000

0 3000

4000 Mass (m/z)

B

SP-C SP-C+5xdodecapeptide

n o i t u o s n i e d i t p e p

3.5 3 2.5 2 1.5 1 0.5 0

100

150

200

Time (h)

Fig. 4. Stability of SP-C with and without SP-Ci N-terminal dodecapeptide. (A) mass spectra of SP-C and a nonaggregating analogue (SP-C33) in 95% aqueous EtOH at initial time (upper lane) and after 120 h (lower lane). The left two spectra show SP-C alone, while the right two spectra correspond to SP-C and ﬁvefold excess of dodecapeptide. The strongest peptide signal in each spectrum is normalized to 100% ion intensity. (B) Amounts of SP-C in solution after different incubation times in 95% aqueous EtOH, determined from peak heights of singly charged SP-C ions in MALDI spectra normalized to the peak height of the internal nonaggregating standard.

FEBS Journal 273 (2006) 926–935 ª 2006 The Authors Journal compilation ª 2006 FEBS

929

J. Li et al.

Properties of SP-C N-terminal propeptide

SP-C alone

A

)

B

SP-C+dodecapeptide A

70 /

60 SP-C+5xdodecapeptide SP-C+2.5xdodecapeptide SP-C

m N m

50

40

30

20

10

200

400

( n o i s n e t e c a f r u S

0

5

10

15

Fig. 5. SP-Ci N-terminal dodecapeptide effect onSP-C ﬁbril forma- tion. (A) SP-C was incubated with a ﬁvefold excess of SP-Ci N-ter- minal dodecapeptide in neutral aqueous organic solvent for 7 days. (B) SP-C alone was incubated under the same conditions.

Time (sec)

B

Instead, amorphous aggregates were detected after the same incubation period.

70 )

/

60 SP-C+5xdodecapeptide SP-C+2.5xdodecapeptide SP-C

m N m

50

Surface activity of SP-C with and without the propeptide

40

30

(POPG),

20

10 ( n o i s n e t e c a f r u S

0

20

40

60

80

100 Surface Area (%)

Fig. 6. Reduced SP-C surface activity in the presence of SP-Ci N-ter- minal dodecapeptide. (A) initial adsorption and (B) surface tension vs. surface area as determined from captive bubble quasi-static cycles. In (B) the ﬁrst and ﬁfth cycles for each mixture are shown.

The surface activity of 2% (w ⁄ w) SP-C in 1,2-dipalmi- toyl-sn-glycero-3-phosphocholine (DPPC) ⁄ 1-palmitoyl- 2-oleoyl-sn-glycero-3-phosphoglycerol 7 : 3 (w ⁄ w), vesicles, and SP-C in the same vesicles plus 2.5- or 5-fold molar excess of the SP-Ci N-terminal dodecapeptide were determined with a captive bubble surfactometer (Fig. 6). The adsorption characteristics of SP-C with a 2.5-fold molar excess of dodecapeptide in phospholipids were similar to those of SP-C alone. In contrast, the mixture with ﬁve times excess of the do- decapeptide showed slow adsorption and did not reach the same surface tension within 5 min as the samples with SP-C alone or SP-C with 2.5 times excess of dodecapeptide. Samples with SP-C alone or SP-C with 2.5-fold excess of dodecapeptide behaved in a similar manner and could be compressed to a minimum surface tension of (cid:1) 0 mnÆm)1 from a maximum surface ten- sion of (cid:1) 25 mnÆm)1. However, a ﬁvefold excess of the dodecapeptide resulted in poor surface activity; the minimum surface tension was (cid:1) 10 mnÆm)1, the maxi- mum surface tension was (cid:1) 50 mnÆm)1, and a larger compression of the surface area was required to reach minimum surface tension.

contributes

to the

is conceivable that

Discussion

The NMR structure of SP-Ci(1–31) in DPC micelles is disordered from Ser1 to Gly11, shows turn and extended structures from residue Arg12 to Phe18, and an a-helix from Pro19 to Leu31 (Fig. 1). This is in agreement with the notion that the region from Ile15 to Phe18 of SP-Ci has a high probalility of forming b-turns [33]. In contrast, SP-C(1–17) in DPC micelles

is disordered from Leu1 to Leu10 and contains an a-helix starting at Lys11 (corresponding to SP-Ci resi- due 23) [31]. This indicates that the presence of the N-terminal dodecapeptide extension in SP-Ci results in structural ordering of the region corresponding to the (cid:1) 10 N-terminal residues of SP-C and also N-ter- minally extends the a-helix four residues. This stabil- signiﬁcantly ization probably longer half-life in neutral solvents of the SP-Ci a-helix the than that of SP-C [29]. It increased stability of SP-Ci, compared to SP-C, is important in order to prevent helix unfolding with concomitant aggregation during the transport and processing of proSP-C in the type II cell. In line with this, the ﬁnal proteolytic processing of SP-Ci to gen- erate SP-C occurs late in the secretory pathway, when SP-Ci is embedded in surfactant phospholipids in the lamellar body [20].

FEBS Journal 273 (2006) 926–935 ª 2006 The Authors Journal compilation ª 2006 FEBS

930

J. Li et al.

Properties of SP-C N-terminal propeptide

Phospholipid vesicles containing SP-Ci have lower surface activity than those with SP-C [28]. A possible explanation for this difference is that the presence of the N-terminal elongation in SP-Ci interferes with the ability of the SP-C N-terminal part to interact with phospholipids. The N-terminal segment of SP-C has intrinsic propensity to interact with phospholipid mem- branes, mediated by both hydrophobic and electro- static interactions [33,36,37]. Adding a ﬁvefold molar excess of the SP-Ci N-terminal dodecapeptide reduces the surface activity of SP-C in DPPC ⁄ POPG (Fig. 6), making it similar to that of SP-Ci in the same lipid mixture [28]. This indicates that SP-C and the dodeca- peptide interact in the presence of phospholipids, as concluded for the same mixture in organic solvents (see above), and that such interactions block the sur- face activity of SP-C.

In conclusion, the properties of a late processing intermediate of proSP-C suggest that the N-terminal propeptide part can regulate the stability and activity of the mature peptide.

Experimental procedures

Peptides

In contrast to the pronounced stability of SP-Ci in neutral solvents, SP-Ci in acidiﬁed solvents undergoes unfolding and formation of amyloid-like ﬁbrils within the same time frame as SP-C [29]. The reduced stability at low pH may in part be explained by the results from the CD experiments of SP-Ci(1–31) in microsomal and surfactant lipids, which show that the peptide undergoes a shift in conformation as the pH is lowered from 6 to 5 (Fig. 3). The CD spectra indicate that this conforma- tional shift is associated mainly with an increase in disordered structure. Since the unpolar a-helix of SP-Ci(1–31) is supposedly embedded in the lipids and the N-terminal end is exposed to the aqueous environ- ment [34], the structural disordering is most likely locali- zed to the N-terminal part. Lamellar bodies are acidic, pH 5.5–6 [30], and it is possible that the pH dependence the conformation of SP-Ci(1–31) now observed of reﬂects that in the early secretory pathway the N-ter- minal part of SP-Ci is folded and the a-helix is stabil- ized, while in the lamellar bodies increased disorder of the N-terminal part facilitates its proteolytic removal. This is in line with the observation that complete pro- cessing to SP-C in lamellar bodies can be inhibited by disruption of the acidic environment [35]. AlaSP-Ci (1–31) incorporated into microsomal or surfactant lipids showed a combination of random and helical structure independent of pH (Fig. 3). This suggests that the N-terminal hexapeptide of SP-Ci (SPPDYS), which is replaced with AAAAAA in AlaSP-Ci(1–31), is import- ant for stabilization of the N-terminal region at neutral pH and for the pH dependent conformational change. It can be noted that Asp4 is the only acidic residue of SP- Ci, and that the PPDYS motif is conserved in all species analysed. The N-terminal region of SP-Ci (residues 10–18 of proSP-C) is important for targeting the protein to distal compartments [21], but whether this function is related to the pH dependent stabilization now observed remains to be investigated.

Isolation of porcine SP-C and surfactant phospholipids

SP-Ci(1–31) (Table 1) was synthesized by solid-phase peptide synthesis using f-Moc chemistry. In SP-Ci(1–31), palmitoyl- Cys at positions 17 and 18 is replaced by Phe, and Val at positions 28–31 is replaced by Leu compared to the wild-type human peptide. Phe was chosen as canine SP-C has one palmitoylCys replaced by Phe [38] and the Val ﬁ Leu replacements were introduced to prevent b-sheet aggregation caused by polyVal [18]. The synthetic SP-Ci(1–31) peptide was dissolved in HAc, followed by dilution with aqueous ethanol to a ﬁnal mixture containing acetic acid ⁄ eth- anol ⁄ water, 3 : 2 : 5 (v ⁄ v). After ﬁltration to remove undis- solved material, the supernatant was applied to a C18 reversed phase HPLC column. SP-Ci(1–31) was eluted with a gradient of acetonitrile ⁄ 0.1% triﬂuoroacetic acid (TFA) run- ning into 0.1% aqueous TFA. Fractions corresponding to SP-Ci(1–31), as detected by MALDI MS, were collected and dried. The dried peptide was stored at )20 (cid:1)C until use. The (SPPDYSAAPRGR) N-terminal dodecapeptide of SP-Ci and AlaSP-Ci(1–31) (Table 1) were from Thermo BioSci- ences GmbH, Ulm, Germany. The SP-C analogue, SP-C33 (Table 1), was synthesized and isolated as described [39].

SP-C can irreversibly transform from its native a-helical structure to b-sheet aggregates and form amyloid-like ﬁbrils both in vivo and in vitro [11–15], while the a-helix of SP-Ci does not unfold in neutral solvents and consequently does not form ﬁbrils [29]. This indicates that the 12-residue elongation at the N-terminal end of SP-Ci stabilizes the helix. Addition of a ﬁvefold molar excess of the N-terminal dodeca- peptide in trans retards SP-C aggregation and prevents the formation of typical amyloid-like ﬁbrils, as judged by electron microscopy (Figs 4 and 5). This suggests that interactions between the dodecapeptide and the N-terminal part of SP-C in solution stabilizes the poly Val helix, and thereby reduces its unfolding and con- comitant aggregation into b-sheets.

FEBS Journal 273 (2006) 926–935 ª 2006 The Authors Journal compilation ª 2006 FEBS

931

SP-C was puriﬁed from the modiﬁed natural surfactant Curosurf (Chiesi Farmaceutici, Parma, Italy), which con- tains phospholipids, SP-B and SP-C. Curosurf paste was

J. Li et al.

Properties of SP-C N-terminal propeptide

nuclear NOESY and TOCSY spectra and analysed with the help of the program xeasy.

CD spectroscopy

The NMR structure of SP-Ci(1–31) is represented by the 20 best conformers calculated with the program dyana [44], and energy minimized using the program opal [45].

MALDI MS

Isolation of microsomal lipids

Two-hundred micromoles of SP-Ci(1–31) and AlaSP-Ci(1– 31) in 0.1 mgÆmL)1 microsome lipids or surfactant lipids were suspended in phosphate buffer and analysed at differ- ent pHs. The CD spectra were recorded, accumulated and averaged in a JASCO J 810 instrument from 190 nm to 260 nm at 22 (cid:1)C using a scan speed of 50 nmÆmin)1, a response time of 2 s, and three spectra per sample. The residual molar ellipticity was calculated after determination of peptide concentration via amino acid analysis. dissolved in methanol ⁄ dichloroethane, 4 : 1 (v ⁄ v), and the phospholipids and hydrophobic proteins were separated by reversed-phase HPLC using a Lipidex 5000 column (40 · 6.5 cm) [40]. Fractions containing SP-B and SP-C were pooled, dried, and redissolved in chloroform ⁄ meth- anol ⁄ 0.1 m hydrochloric acid, 19 : 19 : 2 (v ⁄ v) and SP-C and SP-B were then separated by size-exclusion chromato- graphy on a Sephadex LH60 column (80 · 2.5 cm) [41]. Fractions containing SP-B and SP-C were further puriﬁed by reverse phase HPLC to separate SP-C from SP-B [42]. The samples were applied to a C18 column (Waters), using a ﬂow rate of 0.7 mLÆmin)1. A mixture of 50% aqueous methanol ⁄ 0.1% TFA served as the initial mobile phase and a linear gradient from 0% to 100% of 2-propanol with 0.1% TFA was used for elution. The fractions under the SP-C peak were collected, dried under nitrogen and stored in 95% ethanol at )20 (cid:1)C. MALDI MS conﬁrmed the cor- rect covalent structure of SP-C.

Electron microscopy

Approximately 50 lm SP-C or the same concentration of SP-C with a ﬁvefold molar excess of the SP-Ci N-terminal dodecapeptide SPPDYSAAPRGR were incubated in 95% aqueous ethanol at 22 (cid:1)C. MALDI MS was used for record- ing the aggregation of SP-C by measurements of the rate of disappearance of soluble protein, compared to an internal standard of the nonaggregating SP-C33 [18]. The samples were dried on a stainless steel sample plate containing (cid:1) 5 lg predried a-cyano-4-hydroxycinnamic acid and ana- lysed using a Voyager DePro MALDI TOF (PerSeptive Bio- cystems) instrument operated in the positive ion mode [29].

NMR measurements

liver were minced and rinsed in Five grams of rabbit sucrose. Thereafter, the tissue was homogenized in 10 mm Tris ⁄ HCl pH 7.4 containing 0.25 m sucrose and 1 mm ethy- lenediamine tetra-acetic acid, and the homogenate was cen- trifuged at 20 000 g for 20 min. The supernatant was collected and centrifuged at 100 000 g for 60 min. The pel- let was aliquoted and stored at )80 (cid:1)C until further use. Microsomal lipids were extracted from the 100 000 g pellets by sonication in chloroform ⁄ methanol ⁄ water, 8 : 4 : 3 the meth- (v ⁄ v). After centrifugation and removal of anol ⁄ water phase, the organic phase was washed with chlo- roform ⁄ methanol ⁄ water, 3 : 48 : 47 (v ⁄ v). The combined organic phases were dried under vacuum. samples obtained after

incubation of SP-C or The SP-C ⁄ dodecapeptide in 95% aqueous ethanol were centri- fuged at 20 000 g for 20 min. After removal of superna- tants, the pellets were washed three times with 100 lL water to remove the organic solvent. After ﬁnal centrifuga- tion, pellets were suspended in a small volume of water under low-energy sonication for 5 s. Aliquots of 10 lL of the suspended pellets were placed on grids covered by a formvar ﬁlm. After air-drying, the grids were negatively stained with 1% (w ⁄ v) uranyl acetate in water. The stained grids were examined and photographed in a Philips CM120TWIN electron microscope operated at 80 kV.

Captive bubble surface activity measurements

total measurement

total measurement t2,max ¼ 225 ms,

FEBS Journal 273 (2006) 926–935 ª 2006 The Authors Journal compilation ª 2006 FEBS

932

The lipids DPPC and POPG in a proportion of 7 : 3 (w ⁄ w), were suspended in chloroform ⁄ methanol, 98 : 2 (v ⁄ v). Por- cine SP-C alone, or porcine SP-C together with 2.5- or 5-fold molar excess of the dodecapeptide SPPDYSAAPRGR, were dissolved in chloroform ⁄ methanol, 1 : 1 (v ⁄ v), and added to in 35 mm NMR measurements of 0.5 mm SP-Ci(1–31) DPC, 50 mm sodium phosphate buffer pH 7.5, in 90% H2O ⁄ 10% D2O, were performed at 20 (cid:1)C on a Bruker DMX 600-MHz spectrometer equipped with a cryoprobe, or on a Varian Unity 800 MHz spectrometer. At 800-MHz a NOESY spectrum was recorded (mixing time 40 ms, 4000 · 1800 data points, t1,max ¼ 100 ms, t2,max ¼ 225 ms, total measurement time 72 h). The following spectra were recorded at 600 MHz: NOESY (parameters as above), TOCSY (mixing time 70 ms, 4000 · 1024 data points, t2,max ¼ 225 ms, t1,max ¼ 60 ms, time t1,max ¼ 24 h), and COSY (4000 · 1800 data points, time 48 h). 100 ms, NOESY and TOCSY spectra (recording parameters as above) were recorded at 10 (cid:1)C of a sample of 0.5 mm SP-Ci(1–31) in [2H5]-ethanol, using a Bruker DMX 600- MHz spectrometer equipped with a cryoprobe. Assignments of signals were done in a standard way [43] by using homo-

J. Li et al.

Properties of SP-C N-terminal propeptide

10 Curstedt T, Johansson J, Persson P, Eklund A, Robert- son B, Lowenadler B & Jornvall H (1990) Hydrophobic surfactant-associated polypeptides: SP-C is a lipopeptide with two palmitoylated cysteine residues, whereas SP-B lacks covalently linked fatty acyl groups. Proc Natl Acad Sci USA 87, 2985–2989. 11 Wustneck N, Wustneck R, Perez-Gil J & Pison U

the lipid mixture. The ﬁnal SP-C concentration was 2% by total mass of phospholipids. The mixtures were dried under nitrogen and suspended in saline at a lipid concentration of 10 mgÆmL)1. Surface tension was recorded by a captive bub- ble surfactometer under quasi-static conditions [46]. The air bubble was injected into an airtight chamber containing agarose solution. Lipid ⁄ peptide mixtures were then injected to cover the surface of the air bubble. After an initial adsorp- tion for 5 min, the bubble was compressed stepwise until the minimal surface area was achieved, and then the bubble was expanded stepwise to the original surface area. This process was repeated ﬁve times for each bubble.

(2003) Effects of oligomerization and secondary struc- ture on the surface behavior of pulmonary surfactant proteins SP-B and SP-C. Biophys J 84, 1940–1949. 12 Gustafsson M, Thyberg J, Naslund J, Eliasson E & Johansson J (1999) Amyloid ﬁbril formation by pul- monary surfactant protein C. FEBS Lett 464, 138–142. 13 Szyperski T, Vandenbussche G, Curstedt T, Ruysschaert JM, Wuthrich K & Johansson J (1998) Pulmonary sur- factant-associated polypeptide C in a mixed organic sol- vent transforms from a monomeric alpha-helical state into insoluble beta-sheet aggregates. Protein Sci 7, 2533–2540.

Acknowledgements We are grateful to A˚ ke Engstro¨ m, Uppsala University, for the help with peptide synthesis. This work was supported by the Swedish Research Council and FORMAS.

References

14 Luy B, Diener A, Hummel RP, Sturm E, Ulrich WR & Griesinger C (2004) Structure and potential C-terminal dimerization of a recombinant mutant of surfactant- associated protein C in chloroform ⁄ methanol. Eur J Biochem 271, 2076–2085. 15 Dluhy RA, Shanmukh S, Leapard JB, Kruger P & 1 Johansson J & Curstedt T (1997) Molecular structures and interactions of pulmonary surfactant components. Eur J Biochem 244, 675–693.

2 Goerke J (1998) Pulmonary surfactant: functions and molecular composition. Biochim Biophys Acta 1408, 79–89.

3 Whitsett JA & Weaver TE (2002) Hydrophobic surfac- tant proteins in lung function and disease. N Engl J Med 347, 2141–2148.

Baatz JE (2003) Deacylated pulmonary surfactant pro- tein SP-C transforms from alpha-helical to amyloid ﬁbril structure via a pH-dependent mechanism: an infra- red structural investigation. Biophys J 85, 2417–2429. 16 Gustafsson M, Grifﬁths WJ, Furusjo E & Johansson J (2001) The palmitoyl groups of lung surfactant protein C reduce unfolding into a ﬁbrillogenic intermediate. J Mol Biol 310, 937–950. 17 Kallberg Y, Gustafsson M, Persson B, Thyberg J &

4 Garcia-Verdugo I, Sanchez-Barbero F, Soldau K, Tobias PS & Casals C (2005) Interaction of surfactant protein A (SP-A) with bacterial rough lipopolysaccharide (Re-LPS) and effects of SP-A on the binding of Re-LPS to CD14 and LPS binding protein. Biochem J 391, 115–124. Johansson J (2001) Prediction of amyloid ﬁbril-forming proteins. J Biol Chem 276, 12945–12950. 18 Hosia W, Johansson J & Grifﬁths WJ (2002) 5 Chaby R, Garcia-Verdugo I, Espinassous Q &

Augusto LA (2005) Interactions between LPS and lung surfactant proteins. J Endotoxin Res 11, 181–185. Hydrogen ⁄ deuterium exchange and aggregation of a polyvaline and a polyleucine alpha-helix investigated by matrix-assisted laser desorption ionization mass spectr- ometry. Mol Cell Proteomics 1, 592–597. 6 Senft AP, Korfhagen TR, Whitsett JA, Shapiro SD & 19 Beers MF, Kim CY, Dodia C & Fisher AB (1994)

LeVine AM (2005) Surfactant protein-D regulates solu- ble CD14 through matrix metalloproteinase-12. J Immu- nol 174, 4953–4959. 7 Glasser SW, Korfhagen TR, Bruno MD, Dey C & Localization, synthesis, and processing of surfactant protein SP-C in rat lung analyzed by epitope-speciﬁc antipeptide antibodies. J Biol Chem 269, 20318–20328. 20 Vorbroker DK, Voorhout WF, Weaver TE & Whitsett

Whitsett JA (1990) Structure and expression of the pul- monary surfactant protein SP-C gene in the mouse. J Biol Chem 265, 21986–21991.

FEBS Journal 273 (2006) 926–935 ª 2006 The Authors Journal compilation ª 2006 FEBS

933

JA (1995) Posttranslational processing of surfactant pro- tein C in rat type II cells. Am J Physiol 269, L727–L733. 21 Johnson AL, Braidotti P, Pietra GG, Russo SJ, Kabore A, Wang WJ & Beers MF (2001) Post-translational pro- cessing of surfactant protein-C proprotein: targeting motifs in the NH2-terminal ﬂanking domain are cleaved in late compartments. Am J Respir Cell Mol Biol 24, 253–263. 8 Augusto L, Le Blay K, Auger G, Blanot D & Chaby R (2001) Interaction of bacterial lipopolysaccharide with mouse surfactant protein C inserted into lipid vesicles. Am J Physiol Lung Cell Mol Physiol 281, L776–L785. 9 Johansson J (1998) Structure and properties of surfac- tant protein C. Biochim Biophys Acta 1408, 161–172.

J. Li et al.

Properties of SP-C N-terminal propeptide

22 Conkright JJ, Bridges JP, Na CL, Voorhout WF,

N-terminal segment of pulmonary surfactant protein SP-C from different species. Comp Biochem Physiol A Mol Integr Physiol 129, 129–139. 34 Bi X, Flach CR, Perez-Gil J, Plasencia I, Andreu D, Trapnell B, Glasser SW & Weaver TE (2001) Secretion of surfactant protein C, an integral membrane protein, requires the N-terminal propeptide. J Biol Chem 276, 14658–14664.

Oliveira E & Mendelsohn R (2002) Secondary structure and lipid interactions of the N-terminal segment of pul- monary surfactant SP-C in Langmuir ﬁlms: IR reﬂec- tion-absorption spectroscopy and surface pressure studies. Biochemistry 41, 8385–8395. 35 Beers MF (1996) Inhibition of cellular processing of 23 Nogee LM, Garnier G, Dietz HC, Singer L, Murphy AM, deMello DE & Colten HR (1994) A mutation in the surfactant protein B gene responsible for fatal neo- natal respiratory disease in multiple kindreds. J Clin Invest 93, 1860–1863. 24 Hamvas A, Cole FS, deMello DE, Moxley M, Whitsett

surfactant protein C by drugs affecting intracellular pH gradients. J Biol Chem 271, 14361–14370. 36 Plasencia I, Rivas L, Keough KM, Marsh D &

JA, Colten HR & Nogee LM (1994) Surfactant protein B deﬁciency: antenatal diagnosis and prospective treatment with surfactant replacement. J Pediatr 125, 356–361. 25 Hamvas A, Nogee LM, Mallory GB Jr, Spray TL,

Perez-Gil J (2004) The N-terminal segment of pulmon- ary surfactant lipopeptide SP-C has intrinsic propensity to interact with and perturb phospholipid bilayers. Biochem J 377, 183–193. 37 Plasencia I, Keough KM & Perez-Gil J (2005) Interac-

Huddleston CB, August A, Dehner LP, deMello DE, Moxley M, Nelson R, Cole FS & Colten HR (1997) Lung transplantation for treatment of infants with sur- factant protein B deﬁciency. J Pediatr 130, 231–239. 26 Beers MF, Hamvas A, Moxley MA, Gonzales LW, tion of the N-terminal segment of pulmonary surfactant protein SP-C with interfacial phospholipid ﬁlms. Biochim Biophys Acta 1713, 118–128.

Guttentag SH, Solarin KO, Longmore WJ, Nogee LM & Ballard PL (2000) Pulmonary surfactant metabolism in infants lacking surfactant protein B. Am J Respir Cell Mol Biol 22, 380–391.

38 Johansson J, Persson P, Lowenadler B, Robertson B, Jornvall H & Curstedt T (1991) Canine hydrophobic surfactant polypeptide SP-C. A lipopeptide with one thioester-linked palmitoyl group. FEBS Lett 281, 119–122. 39 Johansson J, Some M, Linderholm BM, Almlen A, 27 Vorbroker DK, Proﬁtt SA, Nogee LM & Whitsett JA (1995) Aberrant processing of surfactant protein C in hereditary SP-B deﬁciency. Am J Physiol 268, L647– L656. 28 Li J, Ikegami M, Na CL, Hamvas A, Espinassous Q,

Curstedt T & Robertson B (2003) A synthetic surfactant based on a poly-Leu SP-C analog and phospholipids: effects on tidal volumes and lung gas volumes in venti- lated immature newborn rabbits. J Appl Physiol 95, 2055–2063. 40 Stark M, Wang Y, Danielsson O, Jornvall H & Chaby R, Nogee LM, Weaver TE & Johansson J (2004) N-terminally extended surfactant protein (SP) C isolated from SP-B-deﬁcient children has reduced surface activity and inhibited lipopolysaccharide binding. Biochemistry 43, 3891–3898. 29 Li J, Hosia W, Hamvas A, Thyberg J, Jornvall H,

Johansson J (1998) Determination of proteins, phos- phatidylethanolamine, and phosphatidylserine in organic solvent extracts of tissue material by analysis of phenylthiocarbamyl derivatives. Anal Biochem 265, 97–102. Weaver TE & Johansson J (2004) The N-terminal pro- peptide of lung surfactant protein C is necessary for biosynthesis and prevents unfolding of a metastable alpha-helix. J Mol Biol 338, 857–862.

30 Chander A (1992) Dicyclohexylcarbodiimide and vana- date sensitive ATPase of lung lamellar bodies. Biochim Biophys Acta 1123, 198–206. 31 Johansson J, Szyperski T & Wuthrich K (1995) Pulmon- 41 Curstedt T, Jornvall H, Robertson B, Bergman T & Berggren P (1987) Two hydrophobic low-molecular- mass protein fractions of pulmonary surfactant. Charac- terization and biophysical activity. Eur J Biochem 168, 255–262.

ary surfactant-associated polypeptide SP-C in lipid micelles: CD studies of intact SP-C and NMR second- ary structure determination of depalmitoyl-SP-C (1–17). FEBS Lett 362, 261–265. 32 Wishart DS, Sykes BD & Richards FM (1992) 42 Gustafsson M, Curstedt T, Jornvall H & Johansson J (1997) Reverse-phase HPLC of the hydrophobic pul- monary surfactant proteins: detection of a surfactant protein C isoform containing Ne-palmitoyl-lysine. Biochem J 326, 799–806. 43 Wuthrich K (1986) NMR of Proteins and Nucleic Acid. Wiley, New York. 44 Guntert P, Mumenthaler C & Wuthrich K (1997) The chemical shift index: a fast and simple method for the assignment of protein secondary structure through NMR spectroscopy. Biochemistry 31, 1647– 1651.

FEBS Journal 273 (2006) 926–935 ª 2006 The Authors Journal compilation ª 2006 FEBS

934

33 Plasencia I, Rivas L, Casals C, Keough KM & Perez- Gil J (2001) Intrinsic structural differences in the Torsion angle dynamics for NMR structure calcula- tion with the new program DYANA. J Mol Biol 273, 283–298.

J. Li et al.

Properties of SP-C N-terminal propeptide

45 Luginbuhl P, Guntert P, Billeter M & Wuthrich K

Supplementary material

The following supplementary material is available online: Table S1. Chemical shifts of SP-Ci(1-31) in DPC micelles. Table S2. Chemical shifts of SP-Ci(1-31) in ethanol.

This material is available as part of the online article

from http://www.blackwell-synergy.com

(1996) The new program OPAL for molecular dynamics simulations and energy reﬁnements of biological macro- molecules. J Biomol NMR 8, 136–146. 46 Schurch S, Bachofen H, Goerke J & Possmayer F

FEBS Journal 273 (2006) 926–935 ª 2006 The Authors Journal compilation ª 2006 FEBS

935

(1989) A captive bubble method reproduces the in situ behavior of lung surfactant monolayers. J Appl Physiol 67, 2389–2396.