Structural behavior of keratin-associated protein 8.1 in human hair as revealed by a monoclonal antibody

Accepted Manuscript Structural behavior of keratin-associated protein 8.1 in human hair as revealed by a monoclonal antibody Hiroki Akiba, Emina Ikeuchi, Ganbat Javkhlan, Hiroki Fujikawa, Osamu AraiKusano, Hiroko Iwanari, Makoto Nakakido, Takao Hamakubo, Yutaka Shimomura, Kouhei Tsumoto PII: DOI: Reference:

S1047-8477(18)30231-4 https://doi.org/10.1016/j.jsb.2018.08.011 YJSBI 7239

To appear in:

Journal of Structural Biology

Received Date: Revised Date: Accepted Date:

2 June 2018 15 August 2018 16 August 2018

Please cite this article as: Akiba, H., Ikeuchi, E., Javkhlan, G., Fujikawa, H., Arai-Kusano, O., Iwanari, H., Nakakido, M., Hamakubo, T., Shimomura, Y., Tsumoto, K., Structural behavior of keratin-associated protein 8.1 in human hair as revealed by a monoclonal antibody, Journal of Structural Biology (2018), doi: https://doi.org/10.1016/j.jsb. 2018.08.011

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Structural behavior of keratin-associated protein 8.1 in human

hair as revealed by a monoclonal antibody

Hiroki Akiba,a,1,2 Emina Ikeuchi,a,1 Ganbat Javkhlan,b Hiroki Fujikawa,c Osamu Arai-Kusano,d Hiroko Iwanari,d Makoto Nakakido,a Takao Hamakubo,d Yutaka Shimomura,c,3 * and Kouhei Tsumotoa,b,e * a

Department of Bioengineering, School of Engineering, The University of Tokyo

b

Department of Chemistry and Biotechnology, School of Engineering, The University of Tokyo

c

Division of Dermatology, Graduate School of Medical and Dental Sciences, Niigata University

d

Laboratory of Quantum Biological Medicine, Research Center for Advanced Science and

Technology, The University of Tokyo e

Medical Proteomics Laboratory, The Institute of Medical Sciences, The University of Tokyo

1

These authors contributed equally to this work.

Current addresses: 2

Laboratory of Pharmacokinetic Optimization, Center for Drug Design Research, National Institutes

of Biomedical Innovation, Health and Nutrition 3

Department of Dermatology, Graduate School of Medicine, Yamaguchi University

1

*

Correspondence

should

be

addressed

to:

[email protected] (Y.S.)

2

[email protected]

(K.T.)

or

Abstract

Keratin-associated protein 8.1 (KAP8.1) is a hair protein whose structure, biochemical roles, and

protein distribution patterns have not been well characterized. In this study, we generated a monoclonal antibody against human KAP8.1 to analyze the protein’s roles and distribution in the

human hair shaft. Using this antibody, we revealed that KAP8.1 was predominantly expressed in

discrete regions of the keratinizing zone of the hair shaft cortex. The protein expression patterns

paralleled the distribution of KAP8.1 mRNA and suggested that KAP8.1 plays a role associated with

cells to control hair curvature. Cross-reactivity among species and epitope analysis indicated that the

monoclonal antibody recognized a linear epitope shared among human, mouse, and sheep KAP8.1.

The antibody failed to interact with sheep KAP8.1in native conformation, suggesting that structural

features of KAP8.1 vary among species.

Keywords

High glycine-tyrosine keratin-associated protein, keratin intermediate filament, immunohistological

staining, surface plasmon resonance, epitope analysis

Abbreviations

BSA, bovine serum albumin; FFF-MALS, field-flow fractionation combined with multi-angle light

3

scattering detector; KAP, keratin-associated protein; HGT KAP, high glycine–tyrosine KAP; HS

KAP, high sulfur KAP; KIF, keratin intermediate filament; MBP, maltose-binding protein; PBS,

phosphate-buffered saline; PBS-T, PBS-Tween 20; SPR, surface plasmon resonance; UHS KAP,

ultra-high sulfur KAP.

Introduction

Keratin-associated proteins (KAPs; the translation products of KRTAP genes) constitute a large family

of proteins produced from at least 100 genes. KAPs influence the formation of the hair shaft through

their interaction with keratin intermediate filaments (KIFs) (Gong et al., 2012; Khan et al., 2014;

Rogers et al., 2006). KIFs are formed within trichocytes and play crucial roles in strengthening the

hair shaft. Morphologically, KIFs are associated each other through a surrounding matrix (Robbins,

2012), of which KAPs are the major component. KAPs are categorized into three groups according to

their amino acid composition, namely, high glycine–tyrosine (HGT), high sulfur (HS), and ultra-high

sulfur (UHS) KAPs (Powell and Rogers, 1997). These proteins are expressed in different regions along

the hair follicle reflecting the differentiation stages (Rogers et al., 2006).

KRTAP genes are shared among mammals, and genetic analysis reveals that those in primates

are highly organized into five clusters generated through gene multiplication (Khan et al., 2014;

Shimomura and Ito, 2005). For example, human HGT KAPs comprise 6 families, the genes for which

4

are all located on chromosome 21q22.1. Both early studies of selective precipitation and gel

electrophoresis of KAP proteins from sheep wool and later analyses of the KRTAP genes have

suggested similar propensities among proteins in the same family and group of KAPs (Rogers et al.,

2006).

Previous studies of the distribution of human KAP mRNAs and proteins showed that different

classes of KAPs are found in different regions within the hair shaft (Rogers et al., 2006; Shimomura

and Ito, 2005). The role of KAPs is believed to be linked to the characteristics of the KIFs in each

region. Despite these analyses, the precise characteristics of specific KAP proteins remain largely

unknown. In particular, only a few analyses of human KAP proteins have been conducted due to the

difficulty of producing and analyzing recombinant proteins with high tyrosine or cysteine content

(Fujikawa et al., 2012; Matsunaga et al., 2013; Rechiche et al., 2018). In addition, molecular

characterization of KAPs remains incomplete because their structures are under complicated tissue-

specific regulation related with their roles in the interaction with KIFs (Fujikawa et al., 2013; Fujimoto

et al., 2014). The interaction of KAPs with KIFs suggests that the control of KAP function might affect

hair shaft formation and modification. Therefore, analysis of the roles of KAPs could lead to new

methods of hair protection and for the treatment of hair-related symptoms, such as alopecia.

Here, we focused on keratin-associated protein 8.1 (KAP8.1), the translation product of the

gene KRTAP8-1 (RefSeq: NP_787053.1). KAP8.1 is the only member of the KAP8 family of HGT

5

KAPs with high Gly (24%) and Tyr (19%) contents. As a member of HGT KAPs, KAP8.1 is suggested

to be strongly expressed in hair cortex (Rogers et al., 2002). For example, in sheep wool, HGT KAPs

are strongly expressed in the orthocortical cells on the convex side of the curl, but are less prevalent

in the paracortical cells on the concave side of the curl (Robbins, 2012). Analysis of a genetic variant

of sheep KAP6.1 has suggested that KAP6.1 plays a a role in controlling wool diameter (Zhou et al.,

2015). Therefore, HGT KAPs appear to contribute to the curvature and strength of the hair shaft by

affecting the packing arrangement of KIFs. Similar investigations have been conducted for human hair

cortical cells, but the results were equivocal (Bryson et al., 2009; Robbins, 2012; Swift, 1997).

Genetic analysis has revealed a more dynamic evolution for HGT KAPs than HS KAPs,

indicating the greater contribution of HGT KAPs in cross-species differences (Wu et al., 2008).

Because the expression patterns of HGT KAP orthologs differ among species (Aoki et al., 1997;

Powell and Rogers, 1997; Rogers et al., 2002), each HGT KAP requires biochemical analysis. Only a

few published reports discuss the biochemical roles of HGT KAPs, among which KAP8.1 is one of

the most studied. Although the distribution of human HGT KAPs has been characterized through in

situ hybridization (Rogers et al., 2002), their behavior at the protein level is unknown, preventing the

development of agents to control the various functions of KAP8.1. Matsunaga et al (2013) used

recombinant proteins to explore the interaction of KAP8.1 with KIFs composed of keratin 85 (K85)

and keratin 35 (K35), and found it to be associated with the head domain of K85. Although this result

6

is consistent with the reported distribution of KRTAP8-1 mRNA in the region K85 is present (Rogers

et al., 2002), no tools to evaluate KAP8.1 protein in hair tissue have been unavailable owing to

difficulties in generating specific antibodies against KAP8.1, which is highly hydrophobic. Here, we

newly generated and analyzed a monoclonal antibody against human KAP8.1, E4304, and used it in

immunohistochemical experiments. Furthermore, we conducted cross-reactivity and epitope analyses

to reveal the characteristics of both the antibody, E4304, and the antigen, KAP8.1.

Material and methods

Purification of KAP8.1 and KAP8.1-His

Human KAP8.1 and KAP8.1-His (human KAP8.1 with a C-terminal hexahistidine tag) were expressed

as described previously and were obtained as inclusion bodies after lysis (Matsunaga et al., 2013).

Other HGT KAPs were cloned and expressed according to the same procedure.

Fourier-transformed infrared spectra

Inclusion bodies of KAP8.1 were ground and mixed with KBr in a standard method. Pellets of KBr

and KBr−KAP8.1 mixture were scanned by using a FT/IR-6300 spectrometer (JASCO) with a 4 cm-1

window and triglycine sulfate detector.

Immunization and cloning of E4304

To conjugate KAP8.1-His with bovine serum albumin (BSA-KAP) for immunization, inclusion bodies

7

containing KAP8.1-His were dissolved in buffer A (8 M urea, 5 mM tris-2-carboxyethylphosphine

hydrochloride [TCEP-HCl], 5 mM imidazole, 5 mM Tris, pH 8.0) and centrifuged at 20,000 × g for

30 min to remove the insoluble fraction. The resulting supernatant was applied to a HisTrap HP

column (GE Healthcare) and washed with buffer A, and then KAP8.1-His was eluted stepwise with

buffer A containing 50, 100, 200, and 300 mM imidazole. The eluate with 300 mM imidazole was

applied to a PD-10 desalting column (GE Healthcare) to exchange its solvent for an 8 M urea.

To

introduce

maleimide

groups

into

BSA,

10

mg/mL

N-(6-

maleimidocaproyloxy)succinimide (Dojindo) in dimethyl sulfoxide and 2 mg/mL BSA

(ThermoFisher) in phosphate-buffered saline (PBS) were mixed at a 40:1 molar ratio and incubated at

room temperature for 90 min. The BSA with maleimide groups (mBSA) was purified over a PD-10

desalting column equilibrated with 8 M urea. KAP8.1-His in 8 M urea and mBSA in 8 M urea were

mixed at a 4:1 molar ratio and concentrated to approximately 2 mg/mL by using an Amicon Ultra-4

centrifugal filter unit (MWCO, 3000; Millipore). The concentrated solution was incubated overnight at 4 ˚C. TCEP-HCl was added to the resulting BSA-KAP solution to achieve a final concentration of

5 mM TCEP-HCl. Female BALB/c mice were immunized by intraperitoneal injection with 100 µg

BSA-KAP in 8 M urea, 5 mM TCEP-HCl mixed with Imject Alum (Thermo Fisher) at a 1:1 ratio, according to manufacturer’s instructions. Immunized mice were boosted 3 times by using the same

inoculum administered at 2-wk intervals. Spleen cells were isolated 3 d after the last immunization

8

and fused with sp2/0-Ag14 mouse myeloma cells through conventional methods (Köhler and Milstein,

1975). Hybridoma culture supernatants were screened by using enzyme-linked immunosorbent assay

with KAP8.1-His–coated 96-well microtiter plates and immunoblotting against KAP8.1-His. The

selected cell was isolated through limiting dilution and used to establish a monoclonal hybridoma cell

line that produced the E4304 antibody against KAP8.1.

Western blotting

To extract whole proteins from a hair sample, several 5-mm strands of cut hair from a healthy Japanese

male were ground with sea sand (Wako Pure Chemical) in a mortar; this mixture was combined with 210 μL homogenization buffer (0.1 M Tris-HCl, 0.05 M dithiothreitol, 2% sodium dodecylsulfate [SDS], 8 M urea, pH 7.5) and 90 μL β-mercaptoethanol with agitation on ice. The mixture was transferred to a 1.5-mL tube, combined with an additional 70 μL homogenization buffer and 30 μL β-mercaptoethanol, and incubated for 1 h at 60 C. The sample was centrifuged at 15,000 × g for 5

min at room temperature to remove the insoluble fraction. The supernatant was mixed directly with

Laemmli sample buffer and separated by using SDS-PAGE. Recombinant proteins were separated by

using essentially the same procedure. Proteins in the gel were transferred to Amersham Nitrocellulose

Western Blotting Membrane (GE Healthcare). The membrane was blocked with 5% skim milk in PBS-

Tween 20 (PBS-T; PBS with 0.05% Tween 20, pH 7.4) for 1 h, and incubated with E4304 antibody

(1/5000) or a monoclonal antibody against maltose-binding protein (MBP) (1/5000; E8032, New

9

England Biolabs) at room temperature for 1 h. The membrane was then washed with PBS-T 3 times,

incubated with goat anti-mouse IgG-HRP (1/5000; sc-2005, Santa Cruz) at room temperature for 1 h,

washed with PBS-T 3 times, and incubated with Amersham ECL Western Blotting Detection Reagents (GE Healthcare) according to the manufacturer’s instructions for detection.

Dot blotting

Samples (2 µL) of each protein in PBS (pH 7.4) were spotted onto a piece of nitrocellulose membrane

and allowed to dry. Blocking, probing, and protein detection were conducted as described for Western

blotting.

Immunohistological analysis

After written informed consent had been obtained, scalp skin samples were harvested from healthy

Japanese individuals (under institutional approval and in adherence to the Declaration of Helsinki

Principles). These samples were either fresh-frozen or fixed in 4% paraformaldehyde to make paraffin blocks. The fresh-frozen scalp skin samples were cut into sections (thickness, 5 μm) by using a

cryostat. The sections were fixed in a 1:1 (v/v) mixture of acetone:methanol for 7 min on ice, blocked in 10% normal goat serum in PBS at room temperature, incubated with E4304 antibody (8.4 μg/mL)

in PBS with 1% normal goat serum at room temperature for 90 min, washed 3 times with PBS,

incubated with goat anti-mouse IgG antibody labeled with AlexaFluor594 (Invitrogen) at room

temperature for 30 min, washed 3 times with PBS, mounted in medium containing 4′,6-diamidino-2-

10

phenylindole for counterstaining, and observed under a fluorescence microscope.

In situ hybridization

A partial cDNA of human KRTAP8-1 (GenBank accession number, NM_175857; nucleotides 233–

441) was PCR-amplified from first-strand cDNA of normal human scalp skin by using primer pairs

reported previously (Rogers et al., 2002) and subsequently cloned into the pCRII-TOPO vector

(Invitrogen). Antisense and sense digoxigenin-labeled cRNA probes were synthesized from linearized

vectors by using T7 and SP6 RNA polymerases (Roche Applied Science), respectively. In situ

hybridization of 4% paraformaldehyde-fixed paraffin sections of normal human scalp skin was

performed as described previously, with minor modifications (Shimomura et al., 2010).

Preparation of MBP-KAP8.1 fusion constructs

Expression vectors for MBP-KAP8.1 were prepared by introducing the gene encoding KAP8.1

(KRTAP8.1) between the NdeI and EcoRI sites of pMAL-c5x (New England Biolabs). MBP-fused

human KAP8.1 peptide fragments were prepared through site-directed deletion of the MBP-KAP8.1

gene by using KOD-Plus Neo (Toyobo). MBP, full-length MBP-KAP8.1, and MBP-KAP8.1 peptides

were prepared by using the BL21(DE3) strain of Escherichia coli as host cells. The soluble fraction

containing MBP or MBP-fused polypeptides were dissolved in buffer B (50 mM Tris-HCl, 200 mM

NaCl, 1 mM EDTA, pH 8.0), sonicated for 10 min, and centrifuged at 40,000 × g for 30 min to remove

the insoluble fraction. The resulting supernatant was applied to amylose resin (New England Biolabs)

11

and washed with buffer B; MBP-fused proteins were eluted by using buffer B containing 10 mM

maltose. The eluate underwent size-exclusion chromatography on a HiLoad 26/60 Superdex 200

column (GE Healthcare) equilibrated with buffer C (50 mM Tris–HCl, 200 mM NaCl, pH 8.0). The

eluted fraction was dialyzed against PBS (pH 7.4) for subsequent analyses.

Field-flow fractionation combined with multi-angle light scattering analysis

The molecular weight of the proteins was determined by using an Eclipse separation system (Wyatt

Technology) equipped with UV and multi-angle light scattering (MALS) detectors. The analysis was conducted as described previously (Kudo et al., 2012). Briefly, 2 μM MBP-KAP8.1 was introduced

into the field-flow fractionation (FFF) system run in buffer C and concentrated by using a focused

flow. Proteins were eluted as the cross flow decreased linearly from 3 to 0 mL/min over 15 min.

Molecular weight was determined according to UV absorbance at 280 nm and MALS data by using

Zimm plotting (ASTRA 5.3, Wyatt Technology).

Surface plasmon resonance

The interaction of E4304 with MBP-KAP8.1 and its fragments was evaluated at 25 °C by surface

plasmon resonance (SPR) (Biacore T200 system, GE Healthcare), with PBS-T as the running buffer.

MBP, MBP-KAP8.1, and E4304 antibody were dialyzed in PBS-T before measurement. To

immobilize E4304 on a sensor chip, the Series S Sensor Chip CM5 and Mouse Antibody Capture Kit (GE Healthcare) were used according to the manufacturer’s instructions. E4304 was diluted to 500

12

nM and then injected into the system at 10 μL/min for 120 s. After 200 s of equilibration, the interaction of E4304 with full-length or truncated MBP-KAP8.1 was detected by washing 1 μM analyte over the chip at 30 μL/min for 120 s, followed by dissociation for 120 s. Kinetic parameters

for the interaction was determined by multi-cycle analysis at various concentrations of the analyte. The surface was regenerated by using 10 mM Glycine-HCl (pH 1.7) at 20 μL/min for 180 s. The cell

flow captured by using anti-mouse antibody was used as the baseline, and curve fitting was conducted

using BIAevaluation Software (GE Healthcare).

Results

Preparation of human KAP8.1 for immunization and the specificity of E4304

Human KAP8.1 and KAP8.1-His (Fig. 1A) were expressed in E. coli and localized entirely in inclusion

bodies. The Fourier-transformed infrared spectrum of the KAP8.1 purified from inclusion bodies

(Supplementary Fig. S1) shared the characteristic absorbance in the amide I region shown previously

for synthetic KAP8.1 peptide (Singh et al., 2017). The protein was soluble at high urea concentrations,

but even 6 M urea was insufficient to maintain it in solution (Supplementary Fig. S2). Because our

attempt to generate antibody by immunizing with urea-solubilized KAP8.1 was unsuccessful, we

chemically conjugated KAP8.1-His to BSA prior and urea-denatured the conjugate prior to

immunization. This conjugate maintained the minimal solubility required for immunization. Western

13

blotting revealed that the resulting monoclonal antibody, E4304, specifically interacted with

recombinant human KAP8.1 in the insoluble fraction (Fig. 1B). This result indicates that E4304

recognized a linear epitope of KAP8.1 without binding to other HGT KAP proteins (e.g., KAP 7.1 and

KAP 19.4). The upper bands likely are dimers and oligomers of KAP8.1 that were not perfectly

disassembled by SDS. The specificity of E4304 for KAP8.1 was further demonstrated through

Western blotting of a whole-protein extract from human hair (Fig. 1C).

Analyses of the localization of KAP8.1 in human hair

For immunohistochemical staining of human KAP8.1, fresh-frozen sections of human scalp were fixed

in methanol–acetone, incubated with E4304, and visualized under a fluorescence microscope (Fig.

2A). In two separate hair follicles, KAP8.1 was expressed predominantly in the keratinizing zone of

the hair shaft cortex, where the hair forms a rigid structure (Fig. 2A). The positive signal was

concentrated in the central portion of the hair shaft cortex (Fig. 2A), consistent with the distribution

of KAP8.1-coding mRNA in the hair shaft as analyzed in parallel through in situ hybridization (Fig.

2B). Taking these results together, we conclude that E4304 specifically detected KAP8.1 protein in

fresh-frozen sections. In contrast, the KAP8.1 signal was less distinct in paraformaldehyde-fixed

sections, even though E4304 recognized a linear epitope. Because KAP8.1 lacks Lys residues, this

result indicates that chemical fixation affected the higher-order structure of KAP8.1, thus shielding

14

the epitope for E4304.

Interaction of E4304 with MBP-fused human KAP8.1 To better understand the KAP8.1 staining patterns in relationship with the protein’s molecular

structure, we analyzed its recognition by E4304 at the molecular level. Here, soluble and homogeneous

KAP8.1 was required; chemical conjugation, which was necessary for immunization with KAP8.1,

often generates a heterogeneous state. Therefore, we sought to introduce a known solubilization tag.

Among those we tested, only the MBP tag enabled the expression of KAP8.1 in the soluble fraction

of the bacterial lysate. Size-exclusion chromatography of MBP-fused human KAP8.1 (MBP-

hKAP8.1) revealed that the fusion protein was monomeric, thus indicating that human KAP8.1 exists

as a monomer rather than a larger assembly (Fig. 3A). Analysis of the monomer fraction by FFF-

MALS (Fig. 3B) indicated that the molecular weight of the main peak was 30.5 kDa, consistent with

that of the MBP-hKAP8.1 monomer calculated from the peptide sequence.

We then used SPR to study the interaction of E4304 with MBP-hKAP8.1 (Fig. 3C). The

sensorgram was successfully fit in 1:1 binding manner with each Fab unit of the antibody showing

dissociation constant of 4.14 nM (Table 1).

Cross-reactivity and epitope analysis of E4304

15

To analyze the cross-reactivity of E4304 among KAP8.1 from different species (Fig. 4A), we used

MBP fusions of mouse, sheep, and alpaca KAP8.1 produced through the same procedure as MBP-

hKAP8.1 (size-exclusion chromatograms are shown in Supplementary Fig. S3). Western blotting

indicated that E4304 bound to the denatured forms of human, mouse, and sheep KAP8.1 but not to

alpaca KAP8.1 (Fig. 4B). Conversely, SPR measurements of the interaction between E4304 and native

KAP8.1 fused to MBP revealed a strong interaction between the antibody and human and mouse

KAP8.1 but not sheep or alpaca KAP8.1 (Fig. 4C; parameters in Table 1 were determined by using

multi-cycle kinetics [Supplementary Fig. S4]). Dot-blots of MBP-fused KAP8.1 probed with anti-

MBP and E4304 yielded a similar result to that provided by SPR in that E4304 detected human and

mouse KAP8.1 only (Supplementary Fig. S5). These combined results strongly indicate that (1) E4304

recognized a peptide sequence shared among human, mouse, and sheep and (2) the tertiary structure

of the region might differ among these species to affect interaction. Epitope determination thus might

provide information regarding the structure of KAP8.1.

To precisely determine the E4304 epitope, we prepared partial peptides of human KAP8.1.

First, full-length KAP8.1 was split into four fragments at Cys residues, and MBP-tagged fusion

proteins were obtained through genetic recombination (Fig. 5A, top). Both Western blotting and SPR

showed that the interaction of E4304 with KAP8.1 was limited to peptide-b (Fig. 5B, C). To further

define the epitope, we divided peptide-b into six fragments (e through j) (Fig. 5A, bottom). Results of

16

Western blotting and SPR clearly showed that peptide-h contained the epitope for E4304 (Fig. 5D, E).

Because shorter fragments within peptide-h (e through g) did not interact (Fig. 5F), the core epitope

of E4304 was identified as peptide-h. The sequence of this region is fully shared among human, mouse,

and sheep KAP8.1, but not alpaca KAP8.1 (Fig. 4A). Therefore, the determined epitope of E4304 was

consistent with its cross-reactivity. When analyzed by SPR, E4304 interacted more tightly with

peptide-b than peptide-h, with contributions in terms of both the association and dissociation steps

(Table 1; see Supplementary Fig. S6 for multi-cycle kinetics). This finding suggests that the C-

terminal half of peptide-b contributed to interaction with KAP8.1 and formed a structure that was

recognized by E4304.

Discussion Our newly developed antibody against human KAP8.1, E4304, recognized both native and denatured

KAP8.1. Immunohistochemistry revealed that the keratinizing zone of the hair shaft cortex stained

heterogeneously (Fig. 2A). This observation supports the fundamental roles of KAP8.1 in the

formation of human hair structure. KAP8.1 interacts with K85 (Matsunaga et al., 2013), which is

widely expressed and distributed around the hair cortex (Langbein et al., 2001). In the current study,

KAP8.1 staining was present only in a well-defined region within the keratinizing zone of the middle

to upper cortex. Therefore, the KAP8.1 protein that we detected in human hair tissue likely does not

17

participate in the initial step of KIF formation, which occurs in the lower matrix, but facilitates the

assembly of already-formed KIFs. That KAP8.1 is expressed in the cortex of the hair shaft is further supported by its mRNA expression pattern, which matched the protein’s localization (Fig. 2B).

Several studies have investigated the distribution of KRTAP mRNAs in human hair tissue

(Rogers et al., 2008, 2007, 2004, 2002, 2001, Shimomura et al., 2002a, 2002b; Soma et al., 2005).

However, because of the difficulties in generating specific antibodies, few studies have assessed the

distribution of human KAP proteins. Among HS KAPs, the KAP1 and KAP2 family proteins are found

in the cortex (Fujikawa et al., 2012; Shimomura et al., 2002b), whereas KAP10 family proteins are

located in the cuticle (Fujikawa et al., 2013). As in the current case of KAP8.1, the distribution of

KAP10 protein matched well with that of the mRNA (Rogers et al., 2004). Among newly determined

KRTAP genes whose amino acid composition differs from those of the three established groups,

KAP24.1 and KAP26.1 showed well-matched protein and mRNA distribution (Rogers et al., 2008,

2007). The present study gave similar but new evidence for the distribution of HGT KAP proteins in

hair by using a specific antibody. Together, the previous and current studies demonstrate the parallel

distribution of KAP proteins and mRNAs in human hair. These findings thus indicate that KAP

proteins are expressed only in specific positions inside the hair shaft and are continuously produced

and degraded according to the differentiation phase. These features contrast with those of keratins,

which are expressed in the lower hair cortex or cuticle and whose proteins are dispersed throughout

18

hair tissue (Langbein et al., 2001, 1999).

Given that the mRNA distribution and protein localization of KAP8.1 matched each other,

the distribution in relationship with the functions of KAP8.1 in hair tissues can be explained by gene

expression. Previously, KRTAP8-1 mRNA was reported to be expressed asymmetrically across the

cortex region, and from the lower to upper cortex, in human beard hair (Rogers et al., 2002). This

pattern differs from our observation of scalp hair, in which KAP8.1 was expressed symmetrically but

heterogeneously in only a limited number of cells in the middle to upper keratinizing zone (Fig. 2A,

B). This difference suggests diversity in the functional mechanisms of KAP8.1 among hair tissues,

conveyed through such factors as the location of the hair (e.g., head hair, eyelash, or beard), sex, age,

and various genetic factors. Interestingly, the expression pattern of KAP8.1 in the present analysis is

consistent with the mRNA distribution of other HGT KAPs (Rogers et al., 2002). These similarities

suggest that HGT KAPs share features of their working mechanisms. In this regard, sheep HGT KAPs

of wool follicles are mainly expressed in orthocortical cells and not in paracortical cells (Powell and

Rogers, 1997); the bilateral distribution of these cells is related to the curvature of the hair tissue. Cells

with similar characteristics with orthocortical cells (classified as type B and type C cells) are present

in human hair follicles (Bryson et al., 2009; Robbins, 2012; Swift, 1997), and straight Japanese hair

demonstrates sparse distribution of type B cells (Bryson et al., 2009). These characteristics are

consistent with our observation that not all cells stained similarly across hair tissues. The presence of

19

different KAP8.1 expression pattern strongly suggests their cell-type–dependent roles in KIF

formation and hair curvature. Although only a few studies have revealed differences in expression

among hair tissues (Rogers et al., 2002), our present results underscore the importance of elucidating

these differences to understand the mechanisms through which KAPs contribute to the formation KIF

assemblies.

How do KAP8.1 molecules promote hair tissue formation as a structural element? Despite

difficulties in solubilizing the KAP8.1 protein produced from E. coli inclusion bodies, we achieved

success through N-terminal MBP fusion, which is considered one of the most effective ways to obtain

soluble fusion proteins (Dyson et al., 2004). According to the size analysis by size-exclusion

chromatography and FFF-MALS, the main product corresponded to MBP-hKAP8.1 monomer.

Consequently, human KAP8.1 would be monomeric under appropriate conditions; this result is

essentially similar to the observation of KAP6.1 in a non-denatured state (Rechiche et al., 2018).

Therefore HGT KAPs are likely to fold and function as monomers. The recently proposed model

structure of KAP8.1 contains two intramolecular disulfide bonds in the absence of an exposed Cys

residue to form dimers or higher-order structures (Singh et al., 2017). This model structure is

consistent with our results.

In the present analyses, we analyzed the interaction of E4304 with KAP8.1. In the model we

described earlier, the epitope region (YWGSYGYPLGYSVG, peptide-b; italics indicate the core

20

epitope in peptide-h) was not fully exposed and extended as a poly-l-proline type II (PPII) helix (note

that Trp16 was substituted to Gly in this figure) (Fig. 6). Therefore, we surmise that structural change

of KAP8.1 is required for interaction with the antibody. The pattern of cross-reactivity among

mammalian species also suggests various characteristics of this region. When the sequences of human,

mouse, and sheep KAP8.1 are compared, the topologic unit between the Cys residues used to generate

peptide-b is the same among these species. However, the sequence corresponding to peptide-c in sheep

KAP8.1 differs from that in the human and mouse orthologs: Cys49 (human and mouse) at the end of

this unit is Ser in sheep; conversely the position corresponding to Tyr55 (human and mouse) is Cys in

sheep. This region is structurally close to the core of the epitope; consequently these amino acid

differences might affect the stability of the structure in solution and thus influence the interaction of

human KAP8.1 with E4304 (Fig. 6). Therefore KAP-specific antibodies might provide a tool for

investigating the structural differences among KAPs of different origins, which may be one of the

factors in the differences between the KIFs formed in these species.

We previously proposed a mechanism for the participation of KAP8.1 in the interaction with K85 or KIFs composed of K35 and K85 through cation-π interaction of the K85 head domain (Arg-

rich) and KAP8.1 (Tyr-rich) (Matsunaga et al., 2013). In the computational model, the Tyr residues

of KAP8.1 are aggregated on the opposite side from the core epitope and at the C-terminus (Fig. 6) (Singh et al., 2017). If this region is tightly bound to KIF by cation-π interaction, then the existence of

21

a relatively flexible region on the other side might be advantageous for the interaction of KAP8.1 with

molecules other than KIFs. Such possibilities are investigated in future experiments, with the goal of

disclosing the dynamic roles of KAP8.1 in the assembly of KIFs, which comprise the major structural

component of hair shafts. These analyses will add to our understanding of the biochemical basis of

hair and ultimately to the development of means to alter its properties.

Conclusion

Our antibody, E4301, specifically interacted with human KAP8.1 in both native and denatured states.

This antibody recognizes a linear epitope, but the higher-order structure of KAP8.1 influenced the

interaction. Using the antibody, we analyzed the distribution of KAP8.1 protein in human hair tissue

and showed that it paralleled the distribution pattern of KAP8.1 mRNA. The current results, combined

with previous studies, highlight the roles of KAP8.1 in specific cells within the keratinizing zone of

the cortex that determine the curvature of the hair tissue. Furthermore, we have revealed that the

conformational states of sheep and human KAP8.1 differ during interaction with the antibody even

though the two species share the same epitope sequence. This difference might reflect differences in

hair tissue between these species and in the association of KAPs with other molecules that influence

the structure of KIFs, which shape hair tissue.

22

Acknowledgements

This work was supported by the Program for World-leading Innovative R&D on Science and

Technology (FIRST) (K.T. and T.H.) and Grants-in-Aid for Scientific Research (nos. JP25249115 and

JP16H02420 to K.T. and 16H06693 to M.N.).

Declarations of interest: none.

References

Aoki, N., Ito, K., Ito, M., 1997. Isolation and Characterization of Mouse High-glycine / Tyrosine Proteins. J. Biol. Chem. 272, 30512–30518. Bryson, W.G., Harland, D.P., Caldwell, J.P., Vernon, J.A., Walls, R.J., Woods, J.L., Nagase, S., Itou, T., Koike, K., 2009. Cortical cell types and intermediate filament arrangements correlate with fiber curvature in Japanese human hair. J. Struct. Biol. 166, 46–58. Dyson, M.R., Shadbolt, S.P., Vincent, K.J., Perera, R.L., McCafferty, J., 2004. Production of soluble mammalian proteins in Escherichia coli: Identification of protein features that correlate with successful expression. BMC Biotechnol. 4, 1–18. Fujikawa, H., Fujimoto, A., Farooq, M., Ito, M., Shimomura, Y., 2013. Characterization of the human hair shaft cuticle-specific keratin-associated protein 10 family. J. Invest. Dermatol. 133, 2780– 2782. Fujikawa, H., Fujimoto, A., Farooq, M., Ito, M., Shimomura, Y., 2012. Characterization of the human hair keratin-associated protein 2 (KRTAP2) gene family. J. Invest. Dermatol. 132, 1806–1813. Fujimoto, S., Takase, T., Kadono, N., Maekubo, K., Hirai, Y., 2014. Krtap11-1, a hair keratin-associated protein, as a possible crucial element for the physical properties of hair shafts. J. Dermatol. Sci. 74, 39–47. Gong, H., Zhou, H., McKenzie, G.W., Yu, Z., Clerens, S., Dyer, J.M., Plowman, J.E., Wright, M.W., Arora, R., Bawden, C.S., Chen, Y., Li, J., Hickford, J.G.H., 2012. An updated nomenclature for keratin-associated proteins (KAPs). Int. J. Biol. Sci. 8, 258–264. Khan, I., Maldonado, E., Vasconcelos, V., O’Brien, S.J., Johnson, W.E., Antunes, A., 2014. Mammalian keratin associated proteins (KRTAPs) subgenomes: Disentangling hair diversity and adaptation to terrestrial and aquatic environments. BMC Genomics 15, 779. Köhler, G., Milstein, C., 1975. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495–497. Kudo, S., Caaveiro, J.M.M., Miyafusa, T., Goda, S., Ishii, K., Matsuura, T., Sudou, Y., Kodama, T., Hamakubo, T., Tsumoto, K., 2012. Structural and thermodynamic characterization of the selfadhesive properties of human P-cadherin. Mol. Biosyst. 8, 2050–2053. Langbein, L., Rogers, M.A., Winter, H., Praetzel, S., Schweizer, J., 2001. The catalog of human hair keratins: II. Expression of the six type II members in the hair follicle and the combined catalog of human type I and II keratins. J. Biol. Chem. 276, 35123–35132. Langbein, L., Rogers, M.A, Winter, H., Praetzel, S., Beckhaus, U., Rackwitz, H., Schweizer, J., 1999.

23

The Catalog of Human Hair Keratins. J. Biol. Chem. 274, 19874–19884. Matsunaga, R., Abe, R., Ishii, D., Watanabe, S. ichi, Kiyoshi, M., Nöcker, B., Tsuchiya, M., Tsumoto, K., 2013. Bidirectional binding property of high glycine-tyrosine keratin-associated protein contributes to the mechanical strength and shape of hair. J. Struct. Biol. 183, 484–494. Powell, B., Rogers, G.E., 1997. The role of keratin proteins and their genes in the growth, structure and properties of hair, in: Formation and Structure of Human Hair. pp. 59–148. Rechiche, O., Plowman, J.E., Harland, D.P., Lee, T.V., Lott, J.S., 2018. Expression and purification of high sulfur and high glycine-tyrosine keratin-associated proteins (KAPs) for biochemical and biophysical characterization. Protein Expr. Purif. 146, 34–44. Robbins, C.R., 2012. Morphological, Macromolecular Structure and Hair Growth, in: Chemical and Physical Behavior of Human Hair. pp. 1–104. Rogers, M.A., Langbein, L., Praetzel-Wunder, S., Giehl, K., 2008. Characterization and expression analysis of the hair keratin associated protein KAP26.1. Br. J. Dermatol. 159, 725–729. Rogers, M.A., Langbein, L., Praetzel-Wunder, S., Winter, H., Schweizer, J., 2006. Human Hair KeratinAssociated Proteins (KAPs). Int. Rev. Cytol. 251, 209–263. Rogers, M.A., Langbein, L., Winter, H., Beckmann, I., Praetzel, S., Schweizer, J., 2004. Hair Keratin Associated Proteins: Characterization of a Second High Sulfur KAP Gene Domain on Human Chromosome 21. J. Invest. Dermatol. 122, 147–158. Rogers, M.A., Langbein, L., Winter, H., Ehmann, C., Praetzel, S., Korn, B., Schweizer, J., 2001. Characterization of a Cluster of Human High/Ultrahigh Sulfur Keratin-associated Protein Genes Embedded in the Type I Keratin Gene Domain on Chromosome 17q12-21. J. Biol. Chem. 276, 19440–19451. Rogers, M.A., Langbein, L., Winter, H., Ehmann, C., Praetzel, S., Schweizer, J., 2002. Characterization of a first domain of human high glycine-tyrosine and high sulfur keratin-associated protein (KAP) genes on chromosome 21q22.1. J. Biol. Chem. 277, 48993–49002. Rogers, M.A., Winter, H., Langbein, L., Wollschläger, A., Praetzel-Wunder, S., Jave-Suarez, L.F., Schweizer, J., 2007. Characterization of human KAP24.1, a cuticular hair keratin-associated protein with unusual amino-acid composition and repeat structure. J. Invest. Dermatol. 127, 1197–1204. Shimomura, Y., Agalliu, D., Vonica, A., Luria, V., Wajid, M., Baumer, A., Belli, S., Petukhova, L., Schinzel, A., Brivanlou, A.H., Barres, B.A., Christiano, A.M., 2010. APCDD1 is a novel Wnt inhibitor mutated in hereditary hypotrichosis simplex. Nature 464, 1043–1047. Shimomura, Y., Aoki, N., Rogers, M.A., Langbein, L., Schweizer, J., Ito, M., 2002a. hKAP1.6 and hKAP1.7, two novel human high sulfur keratin-associated proteins are expressed in the hair follicle cortex. J. Invest. Dermatol. 118, 226–231. Shimomura, Y., Aoki, N., Schweizer, J., Langbein, L., Rogers, M.A., Winter, H., Ito, M., 2002b. Polymorphisms in the human high sulfur hair keratin-associated protein 1, KAP1, gene family. J. Biol. Chem. 277, 45493–45501. Shimomura, Y., Ito, M., 2005. Human Hair Keratin-Associated Proteins. J. Investig. Dermatology Symp. Proc. 10, 230–233. Singh, R.S., Palmer, J.C., Pudney, P.D.A., Paul, P.K.C., Johannessen, C., Debenedetti, P.G., Raut, J., Lee, K., Noro, M., Tiemessen, D., 2017. Molecular modeling and structural characterization of a high glycine–tyrosine hair keratin associated protein. Phys. Chem. Chem. Phys. 19, 8575–8583. Soma, T., Iino, M., Tajima, M., Kishimoto, J., 2005. Expression of novel keratin associated protein 5 genes in the cuticle layer of human hair follicles. J. Dermatol. Sci. 38, 110–112. Swift, J.A., 1997. Morphology and histochemistry of human hair, in: Formation and Structure of Human Hair. pp. 149–175. Wu, D.-D., Irwin, D.M., Zhang, Y.-P., 2008. Molecular evolution of the keratin associated protein gene family in mammals, role in the evolution of mammalian hair. BMC Evol. Biol. 8, 241. Zhou, H., Gong, H., Li, S., Luo, Y., Hickford, J.G.H., 2015. A 57-bp deletion in the ovine KAP6-1 gene affects wool fibre diameter. J. Anim. Breed. Genet. 132, 301–307.

24

Figures and Tables

Fig. 1. A) Diagrams of human KAP8.1 protein constructs. B) Coomassie Brilliant Blue–stained gel

(left) and Western blot (right) of recombinant human HGT KAP proteins expressed in E. coli. M, size

markers; P, whole pellet after lysis; S, whole soluble lysate. C) Western blot of the lysed human hair

sample.

25

Fig. 2. A) Immunohistological images of E4304-stained hair tissue from healthy Japanese donors.

Red: E4304-stained KAP8.1; blue: cell nuclei stained with 4′,6-diamidino-2-phenylindole. B) In situ

hybridization of KRTAP8-1 mRNA in hair samples (upper panel, horizontal section; lower panel,

vertical section). KZ, keratinizing zone; HB, hair bulb.

26

Fig. 3. A) Size-exclusion chromatogram of MBP-hKAP8.1 monitored with absorbance at 280 nm. The

arrow indicates the fraction containing the monomer. B) Molecular weight analyzed by FFF-MALS.

Blue, UV absorbance at 280 nm (right axis); red, calculated molecular weight (left axis). C) SPR

sensorgram of interaction between E4304 and MBP-hKAP8.1.

Table 1. Kinetic parameters in multi-cycle SPR measurements

kon (× 104 M-1s-1)

koff (× 10-4 s-1)

KD (nM)

MBP-hKAP8.1

3.17 ± 0.00

1.31 ± 0.00

4.14 ± 0.02

MBP-mKAP8.1

3.81 ± 0.01

1.75 ± 0.00

4.58 ± 0.01

MBP-peptide-b

1.52 ± 0.00

7.58 ± 0.01

49.9 ± 0.1

MBP-peptide-h

0.521 ± 0.001

0.131 ± 0.000

251 ± 0

hKAP8.1, human KAP8.1; mKAP8.1, mouse KAP8.1. Data are given by the curve fitting values ±

S.E. from multi-cycle kinetics.

27

Fig. 4. A) Homology among the amino acid sequences of human, mouse, sheep, and alpaca KAP8.1

drawn in GENETYX Ver.13 (Genetyx Corporation, Tokyo, Japan). Sequences were extracted from

the NCBI protein database (accession codes: human, Q81UC2.1; mouse, O08633.2; sheep, W5NR85)

or the Ensembl database (alpaca: transcript ID. ENSVPAT00000012459.1). Colors indicate amino

acids of: orange, hydrophobic; blue, negatively charged; purple, positively charged; green, polar;

black, missing. White characters with background colors indicate the amino acids with less homology.

B) Coomassie Brilliant Blue–stained gel (left) and anti-MBP-probed (center) or E4304-probed (right)

Western blot of MBP-fused KAP8.1 from various species expressed in E. coli. M: MBP. h, human;

m: mouse; s, sheep; a, alpaca. C) SPR sensorgrams of the interaction between E4304 and MBP-fused

KAP8.1 of various species. The sensorgram for E4304 with MBP only is shown as a control.

28

Fig. 5. A) Diagram of fragment peptides derived from human KAP8.1. B, D, F) Western blots of each

MBP-fused fragment. Left, anti-MBP; right, E4304. W, MBP-hKAP8.1; M, MBP; a–j, each peptide

with an N-terminal MBP tag. C, E) SPR sensorgrams of each peptide with an N-terminal MBP tag.

29

Fig. 6. Structural elements of human KAP8.1. The model structure was produced by using the

coordinate data presented in Singh et al. (2017) and PyMOL (The PyMOL Molecular Graphics

System). Blue, peptide-h; blue and cyan, peptide-b; yellow, aromatic residues; purple, peptide-c

(whose sequence differs between human and sheep).

30

C)

Cover Image. A) Immunohistological image of E4304-stained hair tissue from healthy Japanese donors. Red: E4304 -stained KAP8.1, blue: DAPI-stained cell nuclei. B) Images of in situ hybridization of KRTAP8-1mRNA in human hair samples. KZ, keratinizing zone; HB, hair bulb. C) Model structure of KAP8.1.

31