Clinical and Experimental Pharmacology and Physiology (2013) 40, 724734

doi: 10.1111/1440-1681.12164

Kidney regeneration by non-platelet RNA-containing particle-derived cells

Wuyi Kong,* Mu Nuo,* Xiao Ping Zhu,* Xiu Juan Han* and Xian Wang
*Beijing Khasar Medical Technology, and Department of Physiology, Beijing University, Beijing, China

SUMMARY
1. We found a group of non-platelet RNA-containing particles (NPRCPs) in human umbilical cord blood. These particles can aggregate, fuse and become non-nucleated cells when
cocultured with nucleated cells in vitro. The non-nucleated
cells further differentiate into nucleated cells expressing octamer binding transcription factor 4 (OCT4). The NPRCPs
are approximately 15 lm in diameter, have a thin bilayer
membrane, contain short RNAs and microRNAs and express
OCT4, sex-determining region Y 2 (SOX2) and DEAD box
polypeptide 4 (DDX4). To conrm the function of NPRCPs in
vivo, we examined the effects of tail vein-injected green uorescent protein (GFP)-labelled NPRCPs on mouse kidneys
damaged by prior ischaemia and reperfusion from Day 1 to
Week 6.
2. Within 1 day of injection of NPRCPs, immunouorescence and immunohistochemistry revealed a large number of
extravasated NPRCPs in the renal calyces, damaged glomeruli and duct tubules. During the course of regeneration,
NPRCPs fused into large, non-nucleated cellular structures
that further became large nucleated cells to regenerate multicellular kidney tubules. In addition, many NPRCPs became
tiny nucleated cellular structures that further differentiated
into interstitial cells in connective tissue. The extravasated
NPRCPs also arranged themselves into non-cell glomerular
structures before further regenerating into nucleated cells of
the glomerulus.
3. In conclusion, the results demonstrate that, via different
patterns of differentiation, NPRCP-derived cells can regenerate mouse kidney tissue damaged by ischaemia.
Key words: acute ischaemia, DEAD box polypeptide 4,
glomerulus, kidney tubules, non-platelet RNA-containing
particles, octamer binding transcription factor 4, sexdetermining region Y 2 , stem cells.

INTRODUCTION
After more than a decade of studies, two unresolved questions
about adult stem cells continue to be debated: (i) whether stem
cells can self-renew; and (ii) what type or types of adult stem
cells possess multilineage regenerative capacity. With regard to
Correspondence: Wuyi Kong, Haidian District, Shangdi E. Street 51,
Jing-Meng A101 Beijing, China. Email: wuyikong@yahoo.com
Received 1 February 2013; revision 30 July 2013; accepted 14 August
2013.
2013 Wiley Publishing Asia Pty Ltd

the rst question, it is believed that stem cell self-renewal occurs

via asymmetric mitotic division and, thus, stem cells provide new
cells for the postnatal lifespan.1 However, the mechanism underlying the asymmetric division of stem cells remains unknown.
We found a group of non-platelet RNA-containing particles
(NPRCPs) in human umbilical cord blood.2 These NPRCPs
express octamer-binding transcription factor 4 (OCT4), sex-determining region Y 2 (SOX2) and DEAD box polypeptide 4
(DDX4) and contain short RNAs and microRNAs. In in vitro
studies,2 in the presence of eukaryotic cells, NPRCPs aggregated,
fused and became a group of non-nucleated cells exhibiting
active movement, including moving in and out of nucleated cells.
These non-nucleated cells further became nucleated OCT4expressing stem cells.
Our ndings provide evidence that new cells can be derived
from recycling, which occurs via the fusion and ssion of
NPRCPs and stem cells. Our data also demonstrate that the formation of new cells is not limited to division by mitosis. Cells
can be formed by the fusion of NPRCPs, or self-assembly. Our
observations support the ndings of Bei et al.3 published
30 years ago, who reported that, during development, cells can
be derived from self-assembly. Bei et al. termed this cell-forming
procedure as cell reformation and claimed that it is a process by
which eukaryotic cells can be reconstructed from non-cellular living material by self-organization.3 Unfortunately, the cell reformation theory was not recognized by others, which was possibly
due to the history of problems within China that disturbed Beis
research.
Another unresolved question regarding stem cells is what type
or types of adult stem cells are multilineage cells. In adults, bone
marrow and blood are important stem cell reservoirs for tissue
regeneration.4,5 Unfractionated bone marrow-derived stem cells
can regenerate muscle, neurons, hepatocytes and smooth muscle
cells69 by differentiation of mesenchymal stem cells
(MSCs).10,11 Although MSC therapy has the potential to establish
a new clinical paradigm, recent trials have reported mixed results,
largely because of the uncertain immunomodulatory properties of
MSCs and the low regenerative efcacy of MSC grafts.1215
Some have suggested that the regeneration of MSCs occurs via a
paracrine effect of released molecules or growth factors and not
as a result of MSC differentiation.16
In the eld of regenerative medicine, non-cellular materials in
the blood have attracted considerable attention. Platelet-rich
plasma (PRP) has been considered to have regenerative
effects.17,18 However, the evidence indicates that the regenerative
effects of platelets in PRP are due to the variety of growth factors
that these platelets release.19 Although there is no direct evidence
supporting a regenerative function of platelets, some have found

NPRCPs regenerate kidney

that platelets can facilitate MSC homing20 and that plateletreleased microparticles promote neural stem cell proliferation and
induce neurogenesis.19,21
In addition to the existence of platelet-related subcellular particles, there are reports that the therapeutic efcacy of MSCs can
be attributed to the factors they secrete, or MSC-released
exosomes. The MSC-released exosomes are approximately 0.1
0.5 lm in diameter and contain small RNAs and premicroRNAs.22 These MSC-released exosomes have been reported to
exhibit a regenerative activity for cardiac cells,23 to have cellular
protective functions in hypoxia-induced pulmonary hypertension24 and to be associated with restored liver function and alleviated liver brosis.25 Proteomic analysis has identied several
hundred proteins; however, OCT4, SOX2 and DDX4 were not
found in MSC-released exosomes.26
In the present study, we examined the process underlying the
reformation of cells by NPRCPs and their regenerative function
in vivo. Our ndings in mice with ischaemia-damaged kidneys
show that NPRCPs assemble into cells in vivo for tissue regeneration.

METHODS
Animals and materials
BALB/c and green uorescent protein (GFP)-transgenic
(FVB.Cg-Tg(ACTB-EGFP)B5Nagy/J) mice (Jackson Laboratory,
Bar Harbor, ME, USA) were bred and maintained in Beijing University. Mice received food and water ad libitum. All animal procedures followed the guidelines of the Animal Regulatory
Department of Beijing University. The OCT4, SOX2, DDX4,
E-cadherin and b1-integrin antibodies were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibodies against b-actin
and tubulin were from CWBiotech (Beijing, China). Alexa Fluor
goat anti-rabbit IgG was from Invitrogen (Carlsbad, CA, USA).
Other chemicals were from Sigma Chemical (Shanghai, China).
Isolation and culture of NPRCPs
Blood was collected from 6 to 8-week-old GFP-transgenic mice
by cardiac puncture. The NPRCPs were isolated from the blood
by centrifugation at 200 g for 10 min at 20C, followed by
centrifugation at 5000 g for 10 min at 20C. The pellet was
resuspended in culture medium (a-minimum essential medium
(MEM) containing 20% fetal bovine serum and antibiotics) on
collagen-coated plates in a 5% CO2 humid incubator at 37C.
The medium was changed every other day. Particles were cultured for >2 weeks before collection for transplantation studies.
Because the lifespan of platelets is <10 days, most contaminating
platelets were removed or were no longer functioning after
2 weeks culture. The cultured NPRCPs were detached by gently
scraping the culture plates with a cell lifter in phosphate-buffered
saline (PBS) containing 0.5 mm EDTA. Detached NPRCPs were
placed in a centrifuge tube and the number of particles determined using a platelet-counting method.27 Briey, 50 lL wellmixed NPRCPs was diluted in 1 mL of 1% ammonium oxalate
in distilled water. The solution was then added to the chamber of
a haemocytometer by capillary action. The haemocytometer was
placed inside a Petri dish for 1020 min to allow the NPRCPs to

725

settle. Then, the NPRCPs were counted under a microscope and

their number calculated according to the dilution factor.
Electron microscopy
Enriched NPRCPs were xed with 2% glutaraldehyde and 4%
paraformaldehyde in sodium cacodylate buffer, pH 7.3, for
30 min at room temperature. After washing with sodium cacodylate buffer, ice-cold 1% osmium tetroxide in distilled water was
added to the pellets, which were gently shaken at 4C for 2 h.
After washing with distilled water, 1% uranyl acetate was added
and samples were incubated overnight at 4C. The pellets were
then dehydrated with serial ethanol dilutions and embedded in
Epon at 65C for 24 h. Ultrathin sections were cut and double
stained with uranyl acetate and lead citrate before the samples
were examined by electron microscopy.
Ischaemia kidney damage and cell transplantation
In all, 54 BALB/c mice were used in the experiments. Eight- to
10-week-old female BALB/c mice were anaesthetized with pentobarbital (40 mg/kg). The abdominal hair was shaved and the
abdominal skin was sterilized with betadine and ethanol before
being opened. Kidney arteries on each side were ligated simultaneously using 40 sutures. After 45 min occlusion, the sutures
were removed to enable blood ow through the arteries again. The
abdominal skin was then closed and 20 million NPRCPs derived
from GFP-transgenic mouse blood were injected into the tail vein
in 150 lL normal saline using a 26 gauge needle. The control
group underwent tail vein injection of 150 lL saline alone. Mice
were killed on Day 1 and Weeks 1, 2, 3, 4 and 6 after transplantation and their kidneys removed and xed for histological processing. Five mice were killed on Day 1 after transplantation and at
least three mice were killed at all other times.
Haematoxylin and eosin, diaminobenzidine and uorescence
staining
The NPRCPs were cultured on collagen-coated cover slides for
3 weeks and were then xed for haematoxylin and eosin (H&E)
staining or immunouorescence studies. Kidneys from normal,
saline-injected or NPRCP-injected mice were collected at different times as described above, xed in 4% paraformaldehyde
overnight and then embedded in parafn. Sections (5 lm) were
cut and washed with PBS before being blocked with blocking
solution containing PBS with 5% horse serum, 5% goat serum
and 0.1% saponine, and then incubated with anti-GFP antibody
overnight at 4C. After overnight incubation, sections were
stained using a VECTASTAIN ABC kit (Vector Laboratories,
Burlingame, CA, USA), counterstained with haematoxylin and
then mounted.
For immunouorescence staining, sections were blocked with
blocking solution and then incubated with different antibodies at
dilutions ranging between 1 : 50 and 1 : 400 in blocking buffer
overnight at 4C. Following overnight incubation, the sections
were washed with blocking solution and incubated with a uorescence-conjugated secondary antibody for 1 h at room temperature. After washing with PBS, sections were counterstained with
4,6-diamidino-2-phenylindole (DAPI) and photographed using

2013 Wiley Publishing Asia Pty Ltd

W Kong et al.

726

conventional uorescence microscopy. Control tissues were treated with non-specic antiserum from the same species of primary
antibody. Staining was visualized by standard uorescence
microscopy (Leica, Wetzlar, Germany).

RESULTS
Morphology and expression of mouse NPRCPs
Non-platelet RNA-containing particles were isolated from GFPtransgenic mice. Two days after seeding, low- and high-magnication microscopy revealed that most populations were small
vesicles of 15 lm in diameter (Fig. 1a,e). These vesicles were
actively moving on the culture plates. Two weeks later, the cultured vesicles were less active. Furthermore, their vesicle structure was replaced by a particle-like structure (Fig. 1b,f). This
morphological change suggests that the NPRCPs may be coated
with serum proteins or that they had absorbed serum proteins
during culture. Non-platelet RNA-containing particles were used
for in vivo transplantation after 2 weeks culture to ensure that the
regenerative function was not via platelets or platelet-released

(k)

(b)

(a)

factors (i.e. because the lifespan of platelets is <10 days, most

platelets in culture would have been removed or would have lost
their function after 2 weeks culture). Histological evaluation of
NPRCPs after 2 weeks culture revealed weak H&E staining
(Fig. 1c,g), so NPRCPs consisted mainly of nuclear materials.
Non-platelet RNA-containing particles express GFP, although
they do not have a known mechanism for protein translation.
Immunouorescent staining conrmed that NPRCPs expressed
GFP (Fig. 1d). Furthermore, double uorescence staining conrmed that most NPRCPs expressed OCT4, DDX4, SOX2,
b-actin and tubulin (Fig. 1ko). The expression of OCT4, SOX2
and DDX4 has not been reported in MSC-released exosomes. In
previous studies using human umbilical cord blood,2 we found
that small NPRCPs result from the ssion of large-sized NPRCPs
or non-nucleated cells that strongly express OCT4. Thus,
NPRCPs are not likely to be enlarged forms of MSC-released
exosomes.
We examined the detailed structure of NPRCPs using electron
microscopy. The particles were found to have different morphological features, which suggests that they are not identical or at
differentiated stages during culture. Some NPRCPs contained

(l)
(c)

(d)

(m)

(e)

(f)

(g)
(n)

(h)

(i)

(j)

(o)

Fig. 1 Morphology and expression of mouse non-platelet RNA-containing particles (NPRCPs), by inverted microscopy, after (a) 2 days and (b) 2 weeks
culture. The NPRCPs were cultured on cover slides for 20 days and were xed and stained with (c) haematoxylin and eosin (H&E) and (d) green uorescent protein (GFP). (eg) High-magnication microscopy of NPRCPs after (e) 2 days and (f) 2 weeks culture and (g) H&E staining. (hj) Electron
microscope images show different morphologies with a dense nuclear structure inside the vesicles. The NPRCPs from GFP-transgenic mice were cultured
for 3 weeks. Immunouorescence staining revealed coexpression of (k) octamer binding transcription factor 4 (OCT4), (l) DEAD box polypeptide 4
(DDX4), (m) sex-determining region Y 2 (SOX2), (n) b-actin and (o) tubulin. Green uorescent protein is expressed on the surface and in the centre of
NPRCPs; OCT4, DDX4, SOX2 are expressed in the centre of NPRCPs and b-actin and tubulin are expressed mainly on the surface of NPRCPs. DAPI,
4,6-diamidino-2-phenylindole. Bars, 20 lm (ad); 10 lm (eg); 500 nm (hj); 10 lm (ko).

2013 Wiley Publishing Asia Pty Ltd

NPRCPs regenerate kidney

large, dense granules (Fig. 1h); others appeared as large dense
bodies (Fig. 1i) and others contained relatively ne granules
(Fig. 1g). All these NPRCPs were approximately 15 lm in
diameter.

(a)

727
(b)

(c)

Migration of NPRCPs to ischaemia-damaged kidneys

Engraftment cell trafcking and homing efcacy are important
for clinical evidence of grafting success. First, we examined the
trafcking of culture-enriched NPRCPs in mouse kidneys 1 day
after tail vein injection. Histological analysis (H&E staining)
revealed that nearly two-thirds of a kidney exhibited ischaemiainduced damage (Fig. 2a). Blood owed from the calyces to the
less-damaged upper kidney, suggesting that the regenerative factors
were carried by the blood circulation to the damaged kidney.
Immunohistochemical and immunouorescence staining
revealed strong GFP staining located in similar areas to that of
blood ow revealed by H&E staining (Fig. 2bd), demonstrating
that a large number of injected NPRCPs migrated to the ischaemia-damaged kidney via blood ow. High-magnication microscopy revealed a large number of short, rod-shaped NPRCPs in
the renal calyces that were positive for GFP (Fig. 2e). These
GFP-positive vesicles aggregated into two groups, which also
contained numerous tiny DAPI-positive dots.
To exclude an autouorescence response of NPRCPs, we also
used immunouorescence staining with a GFP-specic antibody
in saline-injected ischaemia-damaged mouse kidneys. In these
sections, there were no small green vesicles, green vesicle aggregates or large fused green patches at any time point examined.
Sections from control mice did not exhibit any GFP staining in
the calyces (Fig. 2f). At higher magnication, aggregated GFPexpressing NPRCPs detected by uorescence staining were
found to contain numerous short, rod-shaped green vesicles
(Fig. 2g). Diaminobenzidine (DAB) staining with non-specic
antiserum revealed a coloured stain on the extravasated plasma
and negative stains on the aggregated blue dots (Fig. 2h). In
comparison, the use of a GFP-specic antibody revealed coloured stains in extravasated plasma and positive stains in numerous small tiny brown aggregated dots (Fig. 2i). These
observations suggest that the aggregated particles are transplanted NPRCPs that are carried by blood and covered with
plasma (see Fig. S1). However, because plasma readily absorbs
the stain, it is difcult to identify specic stained NPRCPs when
they are covered by plasma (Fig. 2i). Specic staining could
only be conrmed when the plasma was removed. Diaminobenzidine immunohistochemistry revealed dot-like GFP
expression in NPRCPs, whereas immunouorescence staining
revealed vesicle-like GFP expression in NPRCPs. The morphological differences in NPRCPs detected by immunohistochemical
and immunouorescence staining suggest that immunohistochemical staining is less sensitive or that the results are due to dehydration methods used in the DAB immunohistochemistry
procedures. Kidneys from all ve NPRCP-injected mice on Day
1 contained a large number of GFP-expressing NPRCPs. Immunouorescence staining further conrmed that these aggregated
NPRCP were short, rod-shaped vesicles and that some of them
were fused, which suggests their strong plasticity in tissues and
blood ow. Immunouorescence double staining further
conrmed that, similar to NPRCPs in culture, these short,

(d)

(e)

(f)

(g)

(h)

(i)

(j)

(k)

(l)

Fig. 2 Appearance of non-platelet RNA-containing particles (NPRCPs)

in vivo. Green uorescent protein (GFP)-labelled NPRCPs migrated to
ischaemia-damaged mouse kidney 1 day after injection. (a) Haematoxylin
and eosin (H&E) staining and (b) GFPdiaminobenzidine (DAB) staining
of a kidney that also stained for GFP by immunouorescence (c,d)
revealed the location of NPRCPs. (e) Numerous small GFP-expressing
NPRCPs appeared in the renal calyces 1 day after injection in mice with
ischaemia-damaged kidneys. (f) No GFP expression was detected in control kidneys from mice injected with saline. (g) High-magnication
microscopy revealed short, rod-shaped aggregated GFP-positive NPRCPs.
(h,i) Immunohistochemical staining revealed that in the same areas of
ischaemia-damaged mouse kidney found to contain NPRCPs, non-specic
serum stained plasma (h, wide arrow), but not to small blue dots (h, narrow arrow), whereas the GFP-specic antibody stained these small dots
(i, arrow). These ndings were replicated in kidneys from all ve mice
that were injected with NPRCPs. (jl) High-magnication microscopy of
short, rod-shaped GFP-positive NPRCPs coexpressing (j) octamer binding
transcription factor 4 (OCT4), (k) sex-determining region Y 2 (SOX2)
and (l) DEAD box polypeptide 4 (DDX4). Bars, 200 lm (c,d); 20 lm
(e,f); 10 lm (gl).

rod-shaped, GFP-positive NPRCPs coexpressed OCT4, SOX2

and DDX4 (Fig. 2jl), which strongly supports our contention
that these were the injected NPRCPs.

2013 Wiley Publishing Asia Pty Ltd

W Kong et al.

728
Regenerative patterns determined on Day 1 after NPRCP
injection

On Day 1 after NPRCP injection, we examined the distribution

of NPRCPs in engrafted mouse kidneys. The NPRCPs were
found in small groups of aggregates located in the glomerulus
(Fig. 3a), as large aggregates in the tubule area (Fig. 3b) or lined
up in interstitial areas (Fig. 3c). These observations suggest that
the NPRCPs were regenerating different tissues. In addition, a
large number of GFP-positive vesicles was found inside the large
duct tubule (Fig. 3d), further supporting the extravasation of
NPRCPs. Green vesicles were also found lined up in the blood
vessels of the renal capsule (Fig. 3e). Connective tissues near the
calyces of the kidney showed haematoxylin-stained tiny nucleated
particles (Fig. 3f) expressing GFP (Fig. 3g). These tiny particles
were approximately 12 lm in diameter and not uniform in size.
We do not believe these particles were lymphocytes, which usually appear during inammation, because they were much smaller
than lymphocytes. In addition, the aggregated GFP-expressing
NPRCPs coexpressed DDX4, OCT4 and SOX2 (Fig. 3hj). 4,6Diamidino-2-phenylindole staining revealed enormous DAPIpositive particles in the aggregates (Fig. 3j), so the NPRCPs did
not fuse with the nucleated cells but with themselves, supporting
our published in vitro observations.2 Thus, in mice with ischaemia-damaged kidneys, NPRCPs extravasated to different target
tissues immediately after they had arrived via the circulation.
Renal tubule regeneration
In the present study, we rst examined renal tubule regeneration
by NPRCPs. One week after injection, we found no short, rodshaped, GFP-positive particles, but rather large, fused, GFP-positive non-nucleated cellular structures in the kidney (Fig. 4a).
These structures were lined up and located in similar areas where
the aggregated short, rod-shaped NPRCPs were on Day 1 after
injection. These large, GFP-positive, non-nucleated structures had
a tubule shape (Fig. 4b), suggesting NPRCP fusion. Histological
analysis (H&E staining) revealed large cellular structures with
faint haematoxylin staining in the centre of each structure
(Fig. 4c,d). The location and size of these structures was similar
to that of the GFP-positive patches resulting from fusion of
aggregated short, rod-shaped NPRCPs (Fig. 4e). Neither H&E
nor DAPI staining revealed any nuclei in these structures, so the
fusion of NPRCPs was initiated before their nuclear formation.
The large GFP-positive cellular structures coexpressed OCT4 and
differentiated into renal tubules (Fig. 4g), especially into proximal brush border tubules. Both immunouorescence and H&E
staining revealed that the newly derived GFP-positive tubule
structures coexpressed OCT4 (Fig. 4g,h), but no nuclei were
detected. The number of fused green-only patches was lower than
that of fused GFP and red (OCT4) costained patches (Fig. S2),
which suggests that the patches stained red for OCT4 alone were
formed by endogenous NPRCPs. Thus, the regeneration by
NPRCPs differs according to tissue type.
Interstitial cell regeneration
Non-platelet RNA-containing particles that lined up in interstitial
areas (Fig. 3c) differentiated into tiny nucleated cells (Fig. 5).

One week after injection, GFP-positive tiny interstitial cells

appeared in interstitial areas that also stained positive for DDX4;
despite their small size, these interstitial cells stained for tiny
nuclei. These tiny nuclei were easily detected by H&E staining
(Fig. 5). By 2 weeks after injection, GFP staining was not
observed in these interstitial cells, which suggests that these interstitial cells became fully differentiated cells. When new interstitial cells are fully differentiated, GFP is degraded due to the fact
that DNA transcription and normal protein translation have
occurred in these cells. In addition, cellularization of the interstitial tissues occurs earlier than that of renal tubules, suggesting
that the cellularization of interstitial cells requires less fusion and
a shorter incubation time than that of tubule cells.
Glomerulus regeneration
The regeneration of the glomerulus occurred via direct transdifferentiation of NPRCPs. The NPRCPs arranged themselves, but
did not fuse, into capillary net structures before cellularization.
This hypothesis is difcult to prove because it is hard to detect
single NPRCPs because of the fact that non-fused NPRCPs have
weak GFP expression. One week after injection (Fig. 6a), small
particles in a circular area exhibited GFP coexpressed with
DDX4 (Fig. 6a). By 2 weeks after injection, GFP-positive glomeruli were detected, although nuclei were still undetectable
(Fig. 6b). Staining (H&E) revealed the possible regenerative
stages of the glomeruli: rst, NPRCP extravasation (Fig. 6c),
followed by arrangement into a capillary net structure (Fig. 6d)
with eventually the presence of tiny nuclei (Fig. 6e) and later
larger nuclei (Fig. 6f). Cells in the glomerulus still contained
serum protein in cytoplasmic areas (Fig. 6g); the appearance of
the glomerulus further normalized, with erythrocytes in capillaries
(Fig. 6h). These data conrm the cellularization of NPRCPs in
newly regenerated glomeruli.
Kidney regeneration of NPRCPs
Compared with ischaemia-damaged kidneys on Day 1, in which
less than one-third of the kidney was not necrotic (Fig. 2),
approximately two-thirds of ischaemia-damaged mouse kidneys
were not necrotic by 2 weeks after injection of NPRCPs. The
renal cortex regenerated earlier than the renal medulla. Most
ischaemia-damaged kidneys were repaired by 4 weeks after
NPRCP injection (Fig. S3). To conrm that the regeneration was
due to NPRCP-derived cells, low- and high-magnication microscopy was used to examine kidneys 2 weeks after injection of
NPRCPs (Fig. 7). In these kidneys, GFP was expressed in the
unrepaired upper one-third and unrepaired medulla (Fig. 7a,b).
The regenerated areas with normal H&E staining had less GFP
staining (Fig. 7a,b). Higher-magnication microscopy revealed
that extravasated plasma formed glomerulus shapes in the upper
unrepaired damaged areas (Fig. 7c), which were also GFP positive (Fig. 7d). High-magnication microscopy revealed that the
extravasated materials had different morphological features than
nearby erythrocytes (Fig. 7e), so the extravasated materials were
NPRCPs covered with plasma. In addition, GFP staining revealed
that these extravasated materials were non-nucleated GFP-positive
cells that formed glomerulus structures (Fig. 7f) before cellularization to a glomerulus (Fig. 7g). To further conrm that the

2013 Wiley Publishing Asia Pty Ltd

NPRCPs regenerate kidney

(a)

(b)

(d)

(e)

729
(c)

(f)

(g)

(h)

(i)

(j)

Fig. 3 Non-platelet RNA-containing particles (NPRCPs) migrate to damaged kidneys. (ae) One day after injection of NPRCPs in mice with ischaemiadamaged kidneys, NPRCPs were found (a) as small groups of aggregates in a glomerulus (dashed line), (b) large aggregations in the tubule area or (c,d)
lined up in the interstitial areas in the kidneys (c) or in the large tubule (d), as well as (e) in a blood vessel in the renal capsule. (f) Haematoxylin and
eosin (H&E)-stained and (g) green uorescent protein (GFP)-positive tiny nucleated particles were found in connective tissues. (hj) Aggregated NPRCPs
coexpressed (h) DEAD box polypeptide 4 (DDX4), (i) octamer binding transcription factor 4 (OCT4) and (j) sex-determining region Y 2 (SOX2). Tiny
4,6-diamidino-2-phenylindole (DAPI)-positive particles were detected (j). Bars, 20 lm (ad); 10 lm (ej).

2013 Wiley Publishing Asia Pty Ltd

W Kong et al.

730
(a)

(b)

(c)

(d)

(f)

(e)

(g)

Fig. 5 Interstitial tissue regeneration. Tiny cells coexpressed green uorescent protein (GFP; arrows) and DEAD box polypeptide 4 (DDX4) in
interstitial areas 1 week after injection of non-platelet RNA-containing
particles (NPRCPs) into mice with ischaemia-damaged kidneys. Highmagnication microscopy of 4,6-diamidino-2-phenylindole (DAPI; boxed
area) and haematoxylin and eosin (H&E) staining revealed tiny nuclei in
GFP-positive cells (arrows). Bars, 20 lm (merge); 10 lm (H&E).

(h)

Fig. 4 Renal tubule regeneration. Green uorescent protein (GFP)-positive non-platelet RNA-containing particles (NPRCPs) fuse into large
patch-like structures (a) 1 and (b) 2 weeks after injection into mice with
ischaemia-damaged kidneys. (a) The fused patches also coexpress sexdetermining region Y 2 (SOX2; red). (c,d) Haematoxylin and eosin
(H&E) staining revealed large cellular structures containing central weak
staining for nuclear materials (arrows). (e,f) Green uorescent protein uorescence further conrmed that these large cellular structures were
formed by the fusion of NPRCPs (arrows). (g) The GFP-positive large
tubules coexpressed octamer binding transcription factor 4 (OCT4). No
nuclei were detected by (g) 4,6-diamidino-2-phenylindole (DAPI) or (h)
H&E staining. Bars, 100 lm (a); 50 lm (b); 10 lm (ce); 20 lm (fh).

GFP-expressing structures were differentiating into renal tubules,

we examined E-cadherin expression in these structures. One week
after injection of NPRCPs, GFP-expressing cellular structures in
the kidney also stained positive for dot-shaped E-cadherin
(Fig. 7h); 2 weeks after transplantation, the GFP-expressing cellular structures exhibited strong E-cadherin staining. Meanwhile,
nuclei were clearly detected in each GFP-expressing structure
(Fig. 7i), suggesting that cellularization had occurred or nished.
Furthermore, H&E staining revealed large cells in the tubule

structures (Fig. 7j) that were also GFP positive (Fig. 7k). These
data suggest that NPRCPs fused to form new large cells that
further regenerated renal tubules.

DISCUSSION
In the present study we report the presence of NPRCPs in mouse
blood. Similar to NPRCPs in human umbilical cord blood, mouse
NPRCPs express OCT4, SOX2 and DDX4, which has not been
described in the case of MSC-released exosomes. We found that
a large number of injected NPRCPs migrated to ischaemiadamaged kidneys, where they aggregated, fused and became
nucleated stem cells to regenerate the damaged tissues. The
mechanism underlying regeneration by NPRCPs in vivo was
similar to that described in vitro, namely via a self-assembly
mechanism. We also found that most stem cells participating in
the regeneration were derived by self-assembly, which may be
the main mechanism for the production of stem cells by NPRCPs
during tissue damage. Although the circulation contains numerous
small particles, such as exosomes, cellular membrane vesicles,
microparticles and DNA fragments, NPRCPs may not belong to
any of these particle types. The NPRCPs are released from
OCT4-expressing cells,2 possibly via ssion, and are 10-fold
larger than MSC-released exosomes described by others.23

2013 Wiley Publishing Asia Pty Ltd

NPRCPs regenerate kidney

731

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

Fig. 6 Glomerulus regeneration. (a) One week after injection of non-platelet RNA-containing particles (NPRCPs) into mice with ischaemia-damaged
kidneys. Green uorescent protein (GFP) and DEAD box polypeptide 4 (DDX4) are coexpressed as small particles in a non-nuclear area (dashed lines).
(b) Two weeks after transplantation. The GFP-positive glomerulus coexpressed octamer binding transcription factor 4 (OCT4). 4,6-Diamidino-2-phenylindole (DAPI) staining did not reveal any nuclei (dashed lines). (ch) Haematoxylin and eosin (H&E) staining revealed possible regenerative stages of
glomeruli from Day 1 to Week 4 after injection (tiny nuclei with arrows). Bars, 10 lm.

Non-platelet RNA-containing particles are not eukaryotic cells

and therefore do not have a protein translation process. However,
our experiments conrm that NPRCPs express GFP, which can
be detected by a GFP antibody, so NPRCPs contain GFP, but
perhaps not the GFP gene. Green uorescent protein and other
proteins detected in NPRCPs are possibly derived from the ssion
of non-nucleated stem cells. Green uorescent protein, the signature of donors, was no longer expressed in newly derived interstitial cells 1 week after injection and was no longer expressed in
newly derived tubule cells 2 or 3 weeks after injection, possibly
because the new cells initiate mitosis or their own protein translation process. Because these cells do not have the GFP gene, their
GFP expression is temporal. This could also explain why, in previous studies,28 the GFP marker was lost in recipients via an
unknown mechanism after stem cell transplantation. Non-platelet
RNA-containing particles possibly contain regulatory factors for
DNA polymerization and transcription, but not nuclear gene templates. Fused NPRCPs acquire the gene templates from damaged
or adjacent cells in recipients after injection and further differentiate into the same cell lineage that the genes are from.
In the present study, we found that a large number of NPRCPs
migrated to the ischaemia-damaged kidney tissues immediately
after injection and directly regenerated kidney cells. Non-platelet

RNA-containing particles are non-nucleated cells that contain

small RNAs and microRNAs. They regenerate interstitial cells,
renal tubule cells and glomeruli cells by fusion. For interstitial
cell regeneration, NPRCP fusion-derived stem cells use a mechanism similar to that described for stem cell transdifferentiation.
One interstitial cell may be differentiated not from one particle,
but rather the fusion of a few tiny particles. During fusion, the
NPRCPs may fuse some genetic particles released from apoptotic
interstitial cells, which determines the cell lineage. However,
because of the small size of NPRCPs, separating them from
DNA fragments can be difcult. Another explanation for the
regeneration of interstitial cells by NPRCPs is that the NPRCPs
may have several subtypes, which we have identied in in vitro
studies.2 Each NPRCP subtype regenerates cells via a different
mechanism. Non-platelet RNA-containing particles regenerated
the glomerulus also by transdifferentiation. The glomerulus structure is rst formed by the connection of extravasated NPRCPs,
then the cellularization of NPRCPs.
In contrast, a large number of NPRCPs fuse into large cellular
structures that transform into a renal tubule containing many
large-sized cells. Fusion-induced transdifferentiation of stem cells
has been described as stem cells fusing to local cells to acquire
the genetic information, which can occur in vitro on coculture

2013 Wiley Publishing Asia Pty Ltd

W Kong et al.

732
(a)
(b)

(d)

(c)

(e)

(f)

(g)

(h)

(i)

(j)

Fig. 7 Fused non-platelet RNA-containing particles (NPRCPs) become large nucleated stem cells to regenerate renal tubules. (a) Kidneys from NPRCPinjected mice were stained with haematoxylin and eosin and (b) green uorescent protein (GFP). (c,d) In the upper, unrepaired, damaged areas, plasma
forms glomerulus shapes (c, arrows) that are also GFP positive (d, arrows). (e) The materials lling the glomerulus were not erythrocytes (arrow). (f) Glomerulus structures and (g) tubule structures were positive for GFP staining (brown). (h) The GFP-expressing cellular structures also contained dot-shaped
E-cadherin expression (arrows) 1 week after injection of NPRCPs. (i) Large GFP-expressing nucleated cells (arrows) also expressed E-cadherin. (j) Haematoxylin and eosin staining revealed large cells in the tubule structures (arrows) that were also GFP positive (k, arrow). Images are representative results
for kidneys from three mice each at 1 and 2 weeks after injection of NPRCPs transplantation. Bars, 200 lm (c,d); 20 lm (ek).

2013 Wiley Publishing Asia Pty Ltd

NPRCPs regenerate kidney

with epithelial cells and neurons29,30 and in vivo with hepatocytes, intestinal stem cells, cardiomyocytes and neurons after
MSC transplantation.2935 However, fusion of multiple NPRCPs
into a cellular structure may not occur by the same mechanism.
The NPRCPs fuse by themselves to become a cellular structure,
which we observed in our recent in vitro studies.2 The fused
structures contain a large amount of germ cell cytoplasmic components and have extreme plasticity to separate into many nonnucleated cells.2 Large fused patches may obtain genetic information from apoptotic tubule cells, which determines the lineage of
NPRCP-derived tubular cells.
In addition, the nuclear formation of NPRCP-derived cells may
not be through nuclear reprogramming that occurs during nuclear
transfer and cell fusion and with overexpressed stem cell transcription factors in eukaryotic cells.36,37 All these processes
involve using a whole mature nucleus. However, our data provide strong evidence that the nucleus of NPRCP-derived cells is
formed by the fusion of many tiny dot-like nuclear materials, not
a whole nucleus. Fusion-derived large cellular structures become
nucleated cells during differentiation. Nuclear programming
induced by trace amount DNA was described more than 20 years
ago: a small 5 ng amount of phage DNA could form a new
nucleus after injection into Xenopus eggs.38 The assembly of the
mini nuclei does not depend on a specic DNA sequence and
does not require the presence of the egg nucleus, the centromere
or telomere.38 Therefore, de novo nuclear programming can occur
with a small amount of DNA and egg materials are essential for
nucleus formation. We believe that the formation of the nucleus
in NPRCP-derived cells is via a similar mechanism, because
NPRCPs express DDX4 (synonym: VASA), the protein found
specically in germ cells. In addition, NPRCPs express OCT4
and SOX2, the transcription factors for nuclear reprogramming,
which may have some roles in NPRCP cellularization.
The identication of NPRCPs provides strong evidence that
cells can be formed from self-assembly and self-organization. We
believe that stem cells are formed mainly by this mechanism.
Our ndings also support the cell reformation theory introduced
by Beis group 30 years ago.3 Both our in vitro2 and present
studies support the cell reformation theory and demonstrate that
cell reformation is not a rare process: it occurs all the time in our
bodies and most stem cells are derived by this process.
The identication of NPRCPs provides a new approach to cell
therapy. Because they do not have a nucleus, NPRCPs do not
induce rejection and can regenerate into tissue-specic stem cells
after transplantation and thus could be good candidates for cell
therapy.

ACKNOWLEDGEMENTS
This work was supported by a grant from the Department of
Technology, Inner Mongolia Government, China. The authors
thank Laura Smales (BioMedEditing) and Dr Natalie Korszniak
for English language editing.

DISCLOSURE
WK, MN and XZ are employees of Khasar Medicine at Beijing,
China. WK declares the invention of intellectual property led as
NPRCPs, PFDNCs and the use thereof.

733
REFERENCES

1. Morrison SJ, Kimble J. Asymmetric and symmetric stem-cell divisions in development and cancer. Nature 2006; 441: 106874.
2. Kong W, Nuo M, Zhu XP, Han XJ, Luo L, Wang X. Pre-stem cell
formation by non-platelet RNA-containing particle fusion. Clin. Exp.
Pharmacol. Physiol. 2013; 40: 41221.
3. Bei SH, Cao MS, Chen CC et al. Time-lapse microcinematographic
and phase contrast studies on the cell reformation of Chirocephalus
yolk granules in vitro. Sci. Sin. B 1983; 26: 45463.
4. Shizuru JA, Negrin RS, Weissman IL. Hematopoietic stem and progenitor cells: Clinical and preclinical regeneration of the hematolymphoid system. Annu. Rev. Med. 2005; 56: 50938.
5. Wagers AJ, Christensen JL, Weissman IL. Cell fate determination
from stem cells. Gene Ther. 2002; 9: 60612.
6. Ferrari G, Cusella-De Angelis G, Coletta M et al. Muscle regeneration by bone marrow-derived myogenic progenitors. Science 1998;
279: 152830.
7. Mezey E, Chandross KJ, Harta G, Maki RA, McKercher SR. Turning blood into brain: Cells bearing neuronal antigens generated in
vivo from bone marrow. Science 2000; 290: 177982.
8. Lagasse E, Connors H, Al-Dhalimy M et al. Puried hematopoietic
stem cells can differentiate into hepatocytes in vivo. Nat. Med.
2000; 6: 122934.
9. Sata M, Saiura A, Kunisato A et al. Hematopoietic stem cells differentiate into vascular cells that participate in the pathogenesis of
atherosclerosis. Nat. Med. 2002; 8: 4039.
10. Jiang Y, Jahagirdar BN, Reinhardt RL et al. Pluripotency of mesenchymal stem cells derived from adult marrow. Nature 2002; 418: 419.
11. DIppolito G, Diabira S, Howard GA, Menei P, Roos BA, Schiller
PC. Marrow-isolated adult multilineage inducible (MIAMI) cells, a
unique population of postnatal young and old human cells with
extensive expansion and differentiation potential. J. Cell Sci. 2004;
117: 297181.
12. Kebriaei P, Robinson S. Mesenchymal stem cell therapy in the treatment of acute and chronic graft versus host disease. Front. Oncol.
2011; 1: 16.
13. Karp JM, Leng Teo GS. Mesenchymal stem cell homing: The devil
is in the details. Cell Stem Cell 2009; 4: 20616.
14. Ankrum J, Karp JM. Mesenchymal stem cell therapy: Two steps forward, one step back. Trends Mol. Med. 2010; 16: 2039.
15. Phinney DG, Prockop DJ. Concise review: Mesenchymal stem/multipotent stromal cells. The state of transdifferentiation and modes of
tissue repair: Current views. Stem Cells 2007; 25: 2896902.
16. Gnecchi M, Zhang Z, Ni A, Dzau VJ. Paracrine mechanisms in adult
stem cell signaling and therapy. Circ. Res. 2008; 103: 120419.
17. Sanchez-Gonzalez DJ, Mendez-Bolaina E, Trejo-Bahena NI. Platelet-rich plasma peptides: Key for regeneration. Int. J. Pept. 2012;
2012: 532519.
18. Cole BJ, Seroyer ST, Filardo G, Bajaj S, Fortier LA. Platelet-rich
plasma. Where are we now and where are we going?. Sports Health
2010; 2: 20310.
19. Hayon Y, Dashevsky O, Shai E, Varon D, Leker RR. Platelet microparticles promote neural stem cell proliferation, survival and differentiation. J. Mol. Neurosci. 2012; 47: 65965.
20. Jiang L, Song XH, Liu P et al. Platelet-mediated mesenchymal stem
cells homing to the lung reduces monocrotaline-induced rat pulmonary hypertension. Cell Transplant. 2012; 21: 146375.
21. Hayon Y, Dashevsky O, Shai E, Brill A, Varon D, Leker RR. Platelet microparticles induce angiogenesis and neurogenesis after cerebral ischemia. Curr. Neurovasc. Res. 2012; 9: 18592.
22. Chen TS, Lai RC, Lee MM, Choo AB, Lee CN, Lim SK. Mesenchymal stem cell secretes microparticles enriched in pre-microRNAs.
Nucleic Acids Res. 2010; 38: 21524.
23. Lai RC, Chen TS, Lim SK. Mesenchymal stem cell exosome: A
novel stem cell-based therapy for cardiovascular disease. Regen.
Med. 2011; 6: 48192.

2013 Wiley Publishing Asia Pty Ltd

734

W Kong et al.

24. Lee C, Mitsialis SA, Aslam M et al. Exosomes mediate the cytoprotective action of mesenchymal stromal cells on hypoxia-induced pulmonary hypertension. Circulation 2012; 126: 260111.
25. Li T, Yan Y, Wang B et al. Exosomes derived from human umbilical cord mesenchymal stem cells alleviate liver brosis. Stem Cells
Dev. 2013; 22: 84554.
26. Lai RC, Tan SS, Teh BJ et al. Proteolytic potential of the MSC exosome proteome: Implications for an exosome-mediated delivery of
therapeutic proteasome. Int. J. Proteomics 2012; 2012: 971907.
27. Bergmeier W, Rackebrandt K, Schroder W, Zirngibl H, Nieswandt
B. Structural and functional characterization of the mouse von
Willebrand factor receptor GPIb-IX with novel monoclonal antibodies. Blood 2000; 95: 88693.
28. Kong W, Li S, Longaker MT, Lorenz HP. Blood-derived small Dot
cells reduce scar in wound healing. Exp. Cell Res. 2008; 314: 1529
39.
29. Spees JL, Olson SD, Ylostalo J et al. Differentiation, cell fusion,
and nuclear fusion during ex vivo repair of epithelium by human
adult stem cells from bone marrow stroma. Proc. Natl Acad. Sci.
USA 2003; 100: 2397402.
30. Alvarez-Dolado M, Pardal R, Garcia-Verdugo JM et al. Fusion of
bone-marrow-derived cells with Purkinje neurons, cardiomyocytes
and hepatocytes. Nature 2003; 425: 96873.

31. Nygren JM, Jovinge S, Breitbach M et al. Bone marrow-derived

hematopoietic cells generate cardiomyocytes at a low frequency
through cell fusion, but not transdifferentiation. Nat. Med. 2004; 10:
494501.
32. Petersen BE, Bowen WC, Patrene KD et al. Bone marrow as a
potential source of hepatic oval cells. Science 1999; 284: 1168
70.
33. Rizvi AZ, Swain JR, Davies PS et al. Bone marrow-derived cells
fuse with normal and transformed intestinal stem cells. Proc. Natl
Acad. Sci. USA 2006; 103: 63215.
34. Johansson CB, Youssef S, Koleckar K et al. Extensive fusion of haematopoietic cells with Purkinje neurons in response to chronic
inammation. Nat. Cell Biol. 2008; 10: 57583.
35. Vassilopoulos G, Wang PR, Russell DW. Transplanted bone marrow
regenerates liver by cell fusion. Nature 2003; 422: 9014.
36. Gurdon JB, Melton DA. Nuclear reprogramming in cells. Science
2008; 322: 18115.
37. Takahashi K, Yamanaka S. Induction of pluripotent stem cells from
mouse embryonic and adult broblast cultures by dened factors.
Cell 2006; 126: 66376.
38. Forbes DJ, Kirschner MW, Newport JW. Spontaneous formation of
nucleus-like structures around bacteriophage DNA microinjected into
Xenopus eggs. Cell 1983; 34: 1323.

SUPPORTING INFORMATION
Additional Supporting Information may be found in the online version of this article:
Figure S1. Non-platelet RNA-containing particles (NPRCPs) are possibly covered and carried by the circulating plasma. Haematoxylin
and eosin staining shows numerous small dot-shaped particles in the kidney ischaemia-damaged area. Some were covered by plasma
(red) and some initiated cellularization to become larger circled haematoxylin-positive structures (arrows). Bar, 20 lm.
Figure S2. Individual images for the merged images shown in Fig. 4f,g. Aggregated non-platelet RNA-containing particles (NPRCPs)
coexpress DEAD box polypeptide 4 (DDX4), octamer binding transcription factor 4 (OCT4) and sex-determining region Y 2 (SOX2).
Bars, 10 lm.
Figure S3. Kidney regeneration during the 4 weeks after non-platelet RNA-containing particle (NPRCP) transplantation. Ischaemiadamaged kidneys were stained with haematoxylin and eosin. Kidneys 1 day and 1 week after ischaemia exhibit necrotic areas covering
more than two-thirds of the kidney. Two weeks after ischaemic, the necrotic area has decrease to approximately of the kidney. After
3 weeks, the ischaemia-damaged kidney has regenerated the cortex but not the medulla. After 4 weeks, most of the ischaemia-damaged
kidney has been regenerated.

2013 Wiley Publishing Asia Pty Ltd