Ht

MS

a

AA

KDEEHI

1

ttitwbfpt

mgWaueacd(

1

1h

Respiratory Physiology & Neurobiology 189 (2013) 153– 161

Contents lists available at ScienceDirect

Respiratory Physiology & Neurobiology

j ourna l ho me pa ge: www.elsev ier .com/ loca te / resphys io l

aematological acclimation and re-acclimation to hypoxia inhe mouse

eaghan J. MacNutt, James L. Rupert, A. William Sheel ∗

chool of Kinesiology, University of British Columbia, Vancouver, BC, Canada

r t i c l e i n f o

rticle history:ccepted 18 July 2013

eywords:e-acclimation

a b s t r a c t

Haematological responses throughout 4 w of initial acclimation (IA) and three paradigms of re-acclimation (RA) to hypoxia (FIO2 = 0.12) were examined in female mice. We hypothesised that (i)haematological responses would be increased during re-exposure, resulting in greater O2-carryingcapacity in RA compared to IA; and (ii) further improvements would occur when abbreviating the de-

rythropoiesisrythropoietic controlypoxic re-exposure

ntermittent sustained hypoxia

acclimation period to 1 w (RA↓DA) or extending the IA period to 8 w (RA↑IA). The serum [EPO] responsewas blunted in all RA groups compared to IA but the resulting reticulocyte response was similar in allexperimental groups. The [Hb] response was the same in RA and RA↓DA as in IA but was blunted in RA↑IA

due to a reduction in mean corpuscular Hb. The sensitivity of EPO-producing cells appears blunted butthe sensitivity of erythroid precursors to EPO is enhanced by recent hypoxic exposure. Erythropoieticregulation is altered during RA in a manner that is dependent on the paradigm of initial exposure.

© 2013 Elsevier B.V. All rights reserved.

. Introduction

The haematological response to hypoxia represents one ofhe most important features of the acclimation process. Haema-ological acclimation to hypoxia leads to progressive increasesn erythrocyte count (RBC), hematocrit (Hct) and, most impor-antly, haemoglobin concentration ([Hb]). Consequently, alongith increases in the arterial partial pressure of oxygen (PaO2 )

rought about by increased alveolar ventilation, perfusion and dif-usion capacity, arterial content (CaO2 ) is maintained as closely asossible to normoxic levels and O2 delivery to the tissues is pro-ected (Grover and Bärtsch, 1996).

As with many components of the acclimation response, theechanisms underlying haematological adjustments during a sin-

le sustained exposure to hypoxia have been clearly elucidated.ith hypoxia, a rapid increase in [Hb] is brought about by

n immediate hemoconcentration: a reduction in plasma vol-me (PV) and blood volume caused by a shift of fluid from thextracellular to intracellular space (Hannon et al., 1969) and/or

frank diuresis (reviewed in Hoyt and Honig, 1996). The pro-

ess of erythropoiesis is also initiated almost immediately, asemonstrated by a rise in serum erythropoietin concentration[EPO]) within 6 h of hypoxia onset (Knaupp et al., 1992) and

∗ Corresponding author at: 6108, Thunderbird Blvd, Vancouver, BC, Canada V6T-Z3. Tel.: +1 604 822 4459.

E-mail address: bill.sheel@ubc.ca (A. William Sheel).

569-9048/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.resp.2013.07.018

by an increase in circulating reticulocyte count (RC) in as lit-tle as 24 h (Schobersberger et al., 2005). Reticulocytes matureover the following weeks, leading to relatively linear increasesin Hct and [Hb]. When the hypoxic stimulus is removed, haema-tological variables gradually return to normal. The process ofhaematological de-acclimation from hypoxia has been previ-ously described and is mediated by neocytolysis, whereby themost recently formed erythrocytes are targeted for destructionand phagocytised by macrophages in the spleen (Rice et al.,2001).

There is some evidence that both the hemoconcentration (Lyonset al., 1995) and erythropoietic responses (Savourey et al., 1996,2004) to acute hypoxia are altered by previous hypoxic acclimationor acclimatisation; however, the process of haematological accli-mation to sustained re-exposure has received very little attention.In trekkers who were recently monitored throughout two identicaltreks to high altitude (up to 5460 m) separated by 10 d at low alti-tude (1300 m), a significantly greater [Hb] response was observedduring the second exposure (MacNutt et al., 2009, 2012). Given thenumber of potentially confounding factors in such field studies (i.e.diet, exercise, temperature and psychological state), we sought todetermine whether or not these results would be repeatable ina controlled laboratory setting using an animal model. Previousstudies of intermittent hypoxic (IH) exposure in rabbits (6 h d−1

at ∼6100 m for 30 d) demonstrated a greater and more rapidhaematological response during re-acclimation (RA) compared toan initial acclimation (IA) (Jain et al., 1978). However, as there aredifferences between physiological consequences of intermittent

154 M.J. MacNutt et al. / Respiratory Physiology & Neurobiology 189 (2013) 153– 161

IA

RA

RA↓DA

RA↑IA

Time (weeks)

0 1 3 7 14 28d

0 2 4 6 8 10 12 14

0 1 3 7 14 28d

0 1 3 7 14 28d

0 1 3 7 14 28d

F ms ofh ). Anio

vffoatoatDrb2Ddwwibeahahwp

2

2

naoesDvmwiris2cA

ig. 1. Schematic of exposure schedule for initial exposure (IA) and three paradigypoxic exposure and grey shading denotes periods of normoxic de-acclimation (DAf hypoxic exposure or re-exposure.

ersus continuous hypoxia (Sheel and MacNutt, 2008), the resultsrom Jain and colleagues cannot necessarily be extrapolated to dif-erent paradigms of hypoxic re-exposure. For this reason, our firstbjective was to re-address the hypothesis that the haematologicalcclimation to hypoxia (HAH) would occur more rapidly and/oro a greater magnitude during RA compared to IA using a modelf sustained exposure and re-exposure to continuous hypoxia in

controlled laboratory environment. The second objective waso examine the effects of manipulating the duration of IA andA periods on the time course and magnitude of haematological

esponses during RA. Starting with a paradigm that closely resem-led the repeated altitude exposures in trekkers (MacNutt et al.,009, 2012), it was hypothesised that extending IA or abbreviatingA would lead to further facilitation of haematological acclimationuring RA. Rudimentary assessments of haematological statusere employed by Jain et al. (1978) in their work on rabbits asell as in previous field work in humans, offering little insight

nto physiological mechanisms underlying the results reported inoth studies. Thus, the third objective of the current study was toxplore the process of erythropoiesis and test the hypothesis thatlterations in erythropoietic control would account for increasedaematological responses. We hypothesised that: (1) RBC, Hct,nd [Hb] would increase faster and to a greater magnitude duringypoxic re-exposure compared to an initial exposure, and (2) RAould be further facilitated by increasing the duration of the IAeriod and/or decreasing the duration of the DA period.

. Methods

.1. Overview

Experimental animals were initially acclimated (IA) to sustainedormobaric hypoxia, allowed to de-acclimate in normoxia (DA),nd then re-exposed to hypoxia (RA). This protocol was selected inrder to permit comparisons with our human field study (MacNuttt al., 2012). Blood and tissue samples were collected by terminallyampling groups of animals at several time points throughout IA,A and RA. All experimental protocols were approved by the Uni-ersity of British Columbia Animal Care Committee. Adult femaleice (C57BL/6NCrl Charles River Laboratories, Pointe-Claire, QC)ere housed five per cage, with each cage representing an exper-

mental group. C57Bl/6 mice have a very similar haematologicalesponse to hypoxia as two of the other most commonly studied

nbred strains (Balb/c and 129/Sv; Ward et al., 2007) and were cho-en for their resistance to disease and general robustness (Hedrich,004). Animals experienced a 12:12 light:dark cycle and were fedommercial mouse chow ad libitum throughout the experiment.nimals were 10–24 w old and weighed 18–24 g at time of sacrifice.

re-exposure (RA, RA↓DA and RA↑IA) to hypoxia. Black shading denotes periods ofmals in each treatment group were sampled as indicated after 0, 1, 3, 7, 14 and 28 d

2.2. Experimental treatments

2.2.1. Control animalsThree groups of control animals (no hypoxic exposure) were

sampled at ages 10, 12.5 and 20 w to test the effect of age andbody mass on all outcome variables. Since no clear patterns wereseen across this age range for any variables of interest, five animalswere randomly selected from the three age groups to represent thecontrol group. Data from the control group represent baseline (BL)values for each variable.

2.2.2. Initial acclimation to hypoxiaAnimals with no prior hypoxic exposure were sampled through-

out a 4-w IA period after 0 (control group), 1, 3, 7, 14 and 28 d ofhypoxic exposure.

2.2.3. Re-acclimation to hypoxiaAnimals that had been initially acclimated to hypoxia and de-

acclimated in normoxia were sampled after 0, 1, 3, 7, 14 and 28 d ofhypoxic re-exposure. The time domains of IA and DA were manip-ulated to compare three paradigms of haematological RA to IA.

“RA” = RA after 14 d IA and 14 d DA.“RA↓DA” = RA after 14 d IA and an abbreviated (7-d) DA phase.“RA↑IA” = RA after an extended (56-d) IA phase and 14 d DA.

Exposure paradigms for the four experimental groups are illus-trated in Fig. 1. Data were collected from five animals at each of thetime points indicated throughout IA, RA, RA↓DA and RA↑IA.

2.2.4. Hypoxic exposuresUp to eight cages were placed in a plexiglas chamber measuring

51 cm × 71 cm × 36 cm. Hypoxic air was produced using a commer-cially available oxygen extractor (Mountain Air Generator MAG-7,Higher Peak LLC, Winchester, MA, USA) and pumped through thechamber at ∼30 L min−1 to maintain a constant hypoxic environ-ment (FIO2 = 0.12) with negligible CO2 accumulation. During DA,cages were placed on a rack beside the hypoxic chamber. Ambientconditions outside the chamber were monitored regularly and FIO2never dropped below 0.20.

2.2.5. Data collectionAnimals were sacrificed with an overdose of inhaled isoflu-

rane (AErrane®, Baxter Corporation, Mississauga, ON, Canada)after 20 ± 3 min (max 45 min) of removal from the hypoxicchamber.

ology & Neurobiology 189 (2013) 153– 161 155

2

tcftata

2

aepdtoiat(rwsSpa

2

baTsaimw

H

m

r

2

a5aOwcawcBuCa

Table 1Sequences of forward (F) and reverse (R) primers used for amplification of erythro-poietin and �-actin genes.

Gene Primer Sequence (5′–3′)

EPOF: GCT CAG AAG GAA TTG ATG TCG CCTR: ACC CGG AAG AGC TTG CAG AAA GTA

M.J. MacNutt et al. / Respiratory Physi

.2.6. BloodImmediately post mortem, blood was drained by cardiac punc-

ure using a 25 G needle and 1 mL syringe. All blood samples wereollected within 1 min of death. 400 �L of whole blood was trans-erred to an EDTA microtube. Remaining blood (0–500 �L) wasransferred to an SST microtube and allowed to clot at room temper-ture for 30 min. Following centrifugation, serum was transferredo a clean microtube and immediately frozen at −80 ◦C for laterssay of [EPO].

.2.7. KidneysProduction of EPO primarily occurs in the kidney but EPO is

lso expressed in other tissues including the brain. Erythropoi-tin is synthesised in response to hypoxia in endothelial-derivederitubular cells in the kidneys (Lacombe et al., 1988). EPO pro-ucing cells are primarily clustered around proximal tubules ofhe cortex and outer medulla and comprise approximately 10%f the renal interstitial cell population (Koury et al., 1993). EPOs not a stored and released hormone. The response to hypoxia ist the transcriptional level and is driven by hypoxia inducible fac-or (HIF) 1 binding to a promoter site downstream of the EPO geneJelkmann, 2011). In our study, expression of the EPO gene was indi-ectly examined in the kidney. Within 3 min of death, both kidneysere excised, blotted dry and immersed in 1 mL of RNA stabilising

olution (RNAlater®, Applied Biosystems, Streetsville, ON, Canada).amples were held at 4 ◦C for 24 h, allowing solution to thoroughlyermeate the tissue, and were then moved to −80 ◦C for storagend later quantification of EPO mRNA.

.2.8. Complete blood countComplete blood count was obtained from fresh, whole EDTA

lood using an automated low volume veterinary haematologynalyser (Hemavet® 950, Drew Scientific Inc. Dallas, TX, USA).he instrument was calibrated daily to a manufacturer-validatedtandard (MULTI-TROL mouse, Drew Scientific Inc., Dallas, TX, USA)nd re-calibrated after every 25 analyses. Samples were assayedn duplicate. RBC, [Hb] and mean corpuscular volume (MCV) were

easured directly by the instrument and several other variablesere derived as follows:

ct = RBC × MCV10

ean corpuscular Hb (MCH) = 10 ×(

[Hb]RBC

)

ed cell distribution width (RDW)

= 100 ×(

standard deviation of MCVmean MCV

)

.2.9. Reticulocyte countReticulocytes were stained with Thiazole orange and enumer-

ted using a flow cytometric method (Van Hove et al., 1990). Briefly, �L of fresh, whole blood was mixed with 1 mL of a commerciallyvailable stain (BD RETIC-COUNTTM, BD Biosciences, Mississauga,N, Canada). Another 5 �L of blood was mixed with 1 mL of 1× PBSith 0.1% sodium azide solution to serve as an unstained negative

ontrol. Both samples were incubated in the dark at room temper-ture for at least 30 min. Stained samples and unstained controlsere analysed by fluorescence-activated cell sorting using a flow

ytometer equipped with a 488 nm blue argon laser (FACScanTM,

D Biosciences, Mississauga, ON, Canada). Data were collectedsing CellQuestTM software (V3.3, BD Biosciences, Mississauga, ON,anada) and thresholds were adjusted manually to exclude noisend debris. 50,000 events were acquired for each sample and data

�-ActinF: CTG GCT CCT AGC ACC ATG AAG ATCR: TGC TGA TCC ACA TCT GCT GG

were analysed using FlowJo software (V7.6, Tree Star Inc., Ashland,OR, USA). Gating was adjusted around the erythrocyte populationto exclude platelets from analyses. Reticulocytes were identified byoverlaying and subtracting fluorescence histograms for control andstained samples and RC was calculated as the proportion of gatedevents (RBCs) that fluoresced as reticulocytes.

2.2.10. Serum [EPO]A commercially available enzyme-linked immunosorbent assay

kit (Quantikine®, R&D Systems, Minneapolis, MN, USA) was used toquantify serum [EPO]. After a total of 4.5 h incubation, plates wereread at 450 nm with a correction wavelength of 540 nm (BioTekuQuant Universal Microplate Spectrophotometer, Winooski, VT,USA). The mean absorbance of three blank wells was subtractedfrom all sample readings and optical densities (ODs) were con-verted to pg mL−1 using a plate-specific standard curve. Serumsamples were assayed in triplicate and mean values reported.

2.3. Quantification of kidney EPO mRNA

2.3.1. Preparation of cDNAOne kidney from each animal was thawed and homogenised in

1 mL Ambion TRI Reagent® (Applied Biosystems, Streetsville, ON,Canada) on ice. RNA was isolated using a commercially available kit(Ambion RiboPureTM, Applied Biosystems, Streetsville, ON, Canada)and isolated samples were DNAse treated with Ambion® TURBODNA-free (Applied Biosystems, Streetsville, ON, Canada). 1–2 �gof RNA was reverse transcribed in a 10 �L solution using a HighCapacity cDNA Reverse Transcription (RT) kit (Applied Biosystems,Streetsville, ON, Canada). Negative RT (−RT) controls were preparedfor each sample with water in place of reverse transcriptase.

2.3.2. 2.4.3 Quantitative polymerase chain reaction (qPCR)EPO mRNA was quantified using a Stratagene Mx3000P qPCR

system (Agilent Technologies, Mississauga, ON, Canada) and nor-malised to �-actin mRNA, a commonly used internal control(Chan et al., 2010; Li et al., 2005). Primers were designed withPrimerQuestSM software (Integrated DNA Technologies, White-head Institute for Biomedical Research, 2002) using a publishedmouse mRNA sequences (Bult et al., 2008). Sequences of primerpairs for both genes are shown in Table 1. Transcripts were ampli-fied in a 25 �L reaction containing 2 �L of prepared cDNA solutionor −RT control, 400 nmol each of forward and reverse primers,1.25 units of Maxima® Hot Start Taq DNA polymerase (FermentasLife Sciences, Burlington, ON, Canada) and 1.25 �L EvaGreenTM

dye (Biotium, Hayward, CA, USA). A six-point standard curve wasincluded on each plate by preparing fresh dilutions of the samestandard (a mixture of cDNA from several individual mice predictedto have high Epo mRNA based on their exposure). A no-templatecontrol (containing 2 �L water in place of cDNA) was also includedon each plate to detect any DNA contamination of the qPCR mastermix. All reactions were run in triplicate and amplified by cycling

reactions through the following protocol: 10 min at 94 ◦C, 50 cyclesof 15 s at 94 ◦C, 30 s at 58 ◦C (EPO) or 59 ◦C (�-actin) and 45 s at72 ◦C. Upon completion of the reaction, a melting curve analysiswas performed on each qPCR product to confirm the presence of a

1 ology & Neurobiology 189 (2013) 153– 161

suameSppa

2

(lmtdIdRfcnmcttfrpdR

3

3

tadidptpitpiiImwbird

3

geh7

Days in Hypoxia

0 2 4 6 8 10 12 14

Se

rum

[E

PO

] (p

g m

L-1

)

0

100

200

300

400

500

IARARA↓DA

RA↑IA

≠ ●, p < 0.001

Time: p < 0.001 Treatment: p < 0.001 Interaction: p = 0.34**** **

**

**

****

Days in Hypoxi a0 2 4 6 8 10 12 14

Re

lative

EP

O m

RN

A E

xp

ressio

n

0

1

2

3

Time: p = 0.90

Fig. 2. Serum erythropoietin concentration ([EPO]) and kidney Epo mRNA lev-els (normalised to Actb) throughout initial acclimation to hypoxia and threeparadigms of re-acclimation (RA) to hypoxia ([EPO] only). Symbols represent groupmean ± standard error. Horizontal lines represent the mean ± standard error forcontrol animals with no hypoxic exposure. Results of two-way ANOVA and sig-nificant differences between mean IA and RA responses (Dunnett’s test) are shown

significantly different from IA (all p < 0.001). Although peak RC wasconsistent across treatment groups at 7–8% after 7 d, RC was lowerin all RA groups than IA after 3 d and 14 d in hypoxia.

0 1 2 3 4 5 6

[EP

O] (p

g m

L-1

)

0

200

400

600

r = 0.78

n = 17

p < 0.001

56 M.J. MacNutt et al. / Respiratory Physi

ingle amplification product. No EPO or �-actin amplication prod-ct was detected in any −RT or no-template control, indicating

lack of genomic DNA contamination in cDNA samples or qPCRaster mix. Critical amplification thresholds were determined for

ach reaction using MxPro software (Mx 3000P version 4.10, ©2007tratagene, Agilent Technologies, Mississauga, ON). For each sam-le, the quantity of initial transcript was calculated relative to thelate-specific standard curve and normalised to �-actin mRNA toccount for inter-sample differences in cDNA content.

.3.3. Data analysisAll analyses were completed using a statistical software package

SPSS 15.0, IBM SPSS Statistics, Chicago, USA) and the significanceevel of all tests was set at ̨ = 0.05. Data are presented as group

ean ± standard error. Haematological status immediately prioro re-exposure was compared to BL using unpaired t-tests withata collected from mice sampled at IA14-DA14, IA14-DA7 and

A56-DA14. A two-way analysis of variance (ANOVA) was used toetermine the effect of time in hypoxia (0–28 d) and treatment (IA,A, RA↓DA and RA↑IA) on each outcome variable. If the F statistic

or treatment reached significance, the three RA treatments wereompared to IA using a Dunnett’s test. In cases where data didot meet the assumption of homogeneity of variances (as deter-ined by Levene’s test), the Dunnet’s T3 test was used for post hoc

omparison of each RA treatment to IA. The Tukey’s HSD statis-ic was used to examine post hoc differences between IA and thehree RA treatments at a given time point. Serum was availableor two to four animals per group. Group means were used toeplace missing values so analyses for serum [EPO] could be com-leted as above with n = 4. Due to technical issues, MCV and itserived variables (Hct and RDW) are unavailable for RA↓DA andA↑IA groups.

. Results

.1. [EPO] and Epo mRNA

The [EPO] was significantly affected by both time in hypoxia andreatment group (shown in Fig. 2). In IA, [EPO] increased rapidlynd peaked after 3 d at more than double its initial value beforeecreasing back towards, and below, BL. [EPO] was still below BL

mmediately before all three re-exposure paradigms, though theifference from BL was significant only for RA↓DA and RA↑IA. Theattern of [EPO] response was similar across all treatments, buthe magnitude was notably reduced in all three RA groups com-ared to IA. In RA, peak [EPO] was reached after 1 d, instead of 3 d

n hypoxia and although the peak was more than 100 pg mL−1 lowerhan in IA this did not represent a significant reduction. Conversely,eak [EPO] was significantly lower in both RA↓DA and RA↑IA than

n IA. The overall [EPO] response was most notably blunted dur-ng RA↑IA and this is the only group significantly different fromA. We observed no statistical effect of time in hypoxia on Epo

RNA. However, among the 17 animals for which both types of dataere available, there existed a strong correlation (r = 0.78, p < 0.001)

etween individual Epo mRNA and EPO protein (see Fig. 3), suggest-ng that the intra-group variability in the quantity of Epo mRNAeflects physiological differences and not methodological or ran-om error.

.2. Reticulocyte count

RC was significantly affected by time in hypoxia and treatment

roup, but there was also a significant interaction between the mainffects (all p < 0.001, see Fig. 4). In IA, RC increased after 1 and 3 d inypoxia then plateaued until measurements stopped at 14 d. After

d DA, RC had dropped below BL levels but was back to normal after

at the bottom of the panel. Significant differences between IA and RA at each timepoint (Tukey’s HSD) are marked with **(p < 0.01).

14 d DA. Post hoc tests revealed that all three RA conditions were

Epo mRNA (arbitrary units)

Fig. 3. Scatter plot of serum [EPO] versus normalised Epo mRNA for 17 mice withavailable data. The Pearson product moment correlation and corresponding p valueare given.

M.J. MacNutt et al. / Respiratory Physiology & Neurobiology 189 (2013) 153– 161 157

Fig. 4. Reticulocyte count during an initial acclimation (IA) and three paradigms ofre-acclimation (RA) to hypoxia. Horizontal lines represent the mean ± standard errorfor control animals with no hypoxic exposure. Dashed lines connect non-adjacenttime points when data are missing. Results of two-way ANOVA and significant differ-eow

3

hieHbnrtmrtphrrt

tIilga1

RaRtrie

3

ttb

Days in Hypoxia

0 5 10 15 20 25 30

RB

C (

M u

L-1

)

0

8

9

10

11

12

Days in Hypoxia

0 5 10 15 20 25 30

[Hb] (g

dL

-1)

0

12

13

14

15

16

17

Days in Hypoxia

0 5 10 15 20 25 30

MC

H (

pg

)

0.0

13.5

14.0

14.5

15.0

15.5

16.0

16.5

≠ ●

● ●

, p = 0.005

≠ , p = 0.01; ≠ , p < 0.001

**

***

***

** ** ** **

Time: p < 0.001 Treatment: p = 0.09 Interact ion: p = 0.003

Time: p < 0.001 Treatment: p < 0.001 Interact ion: p = 0.001

Time: p < 0.001 Treatment: p < 0.001 Interaction: p < 0.001

IARARA↓DA

RA↑IA

**

Fig. 5. Erythrocyte count (RBC), haemoglobin concentration ([Hb]) and mean cor-puscular haemoglobin (MCH) during an initial acclimation (IA) and three paradigmsof re-acclimation (RA) to hypoxia. Symbols represent group mean ± standard error.Horizontal lines represent the mean ± standard error for control animals with nohypoxic exposure. Results of two-way ANOVA and significant differences betweenmean IA and RA responses (Dunnett’s test) are shown at the bottom of each panel.

nces between mean IA and RA responses (Dunnett’s test) are shown at the bottomf the panel. Significant differences between IA and RA at each time point are markedith *(p < 0.05) or **(p < 0.01).

.3. RBC, [Hb] and MCH

RBC, [Hb] and MCH were significantly affected by time inypoxia (all p < 0.001). Although [Hb] and MCH were also signif-

cantly affected by treatment group (both p < 0.001), the treatmentffect only approached statistical significance for RBC (p = 0.09).owever, for all three variables there was a significant interactionetween time and treatment, indicating a need for closer exami-ation of response patterns for each variable (see Fig. 5). The RBCesponse was consistent between treatment groups and Dunnett’sest found no significant difference between IA and any RA treat-

ent. Although not significantly different, RBC had not completelyeturned to BL before re-exposure in RA↑IA and RBC in this groupended to be higher throughout the first week of exposure beforelateauing at a similar level to IA and RA after 14 and 28 d inypoxia. Conversely, although de-acclimation was complete beforee-exposure and the early response was identical to IA and RA, RBCeached a significantly higher peak after 28 d in hypoxia in RA↓DAhan in IA.

Although [Hb] was highly variable at the end of 28 d in hypoxia,he [Hb] response was consistent throughout IA, RA and RA↓DA.n RA↑IA, [Hb] remained slightly, but not significantly, above BLmmediately before re-exposure. However, [Hb] increased to aesser degree throughout RA↑IA as compared to the other treatmentroups. Dunnett’s test indicated a significant difference between IAnd RA↑IA, with [Hb] significantly lower in RA↑IA than IA at 3 d and4 d.

MCH was significantly elevated before re-exposure in RA andA↓DA, but the overall MCH response was not different between IAnd RA. Dunnett’s test indicated a significant difference betweenA↓DA and IA, though this was driven largely by differences inhe early MCH response. Following an extended IA period, MCHeturned to BL after 2 w DA. However, MCH was significantly lowern RA↑IA than IA with MCH consistently below BL throughout thexposure.

.4. MCV, Hct and RDW

MCV, Hct and RDW data were available only for IA and RAreatments (see Fig. 6). There was a significant effect of time andreatment for all three variables, as well as a significant interactionetween the main effects (all p < 0.001).

Significant differences between IA and RA at each time point (Tukey’s HSD) aremarked with *(p < 0.05) or **(p < 0.01).

MCV increased throughout IA and returned to BL before re-exposure. MCV also increased throughout RA, but to a lesser extentthan in IA. Mean cell size was significantly smaller in RA thanIA after 7, 14 and 28 d in hypoxia. Hct followed a similar pat-tern, increasing throughout IA and de-acclimating completely toBL before re-exposure. Again, Hct increased throughout RA but to alesser degree than in IA. However, there were no significant differ-ences in Hct between IA and RA at any time point. RDW increased

sharply during IA, peaking at 7 d then gradually returning towardsBL. However, de-acclimation was not complete after 14 d in nor-moxia and RDW remained significantly greater than in IA at 0 and

158 M.J. MacNutt et al. / Respiratory Physiology &

Days in Hypoxia

0 5 10 15 20 25 30

MC

V (

fL)

0

44

46

48

50

52 IARA

Days in Hypoxia

0 5 10 15 20 25 30

Hct (%

)

0

35

40

45

50

55

Days in Hypoxia

0 5 10 15 20 25 30

RD

W (

%)

0

16

17

18

19

20

**

**

**

**

**

Time: p < 0.001 Treatment: p < 0.001 Interaction: p = 0.001

Time: p < 0.001 Treatment: p < 0.001 Interaction: p < 0.001

Time: p < 0.001 Treatment: p < 0.001 Interact ion: p < 0.001

Fig. 6. Mean corpuscular volume (MCV), haematocrit (Hct) and red cell distri-bution width (RDW) during an initial acclimation (IA) and three paradigms ofre-acclimation (RA) to hypoxia. Symbols represent group mean ± standard error.Horizontal lines represent the mean ± standard error for control animals with nohpa

3t

4

cd(tI

ypoxic exposure. Results of two-way ANOVA are shown at the bottom of eachanel. Significant differences between IA and RA at each time point (Tukey’s HSD)re marked with **(p < 0.01).

d in RA. As in IA, RDW peaked at 7 d in RA then steadily decreasedowards BL over the remaining weeks in hypoxia.

. Discussion

It was hypothesised that: (1) RBC, Hct, [Hb], and therefore O2arrying capacity, would increase faster and to a greater magnitude

uring hypoxic re-exposure compared to an initial exposure, and2) RA would be further facilitated by increasing the duration ofhe IA period and/or decreasing the duration of the DA period.n order to explore the mechanisms of facilitated haematological

Neurobiology 189 (2013) 153– 161

re-acclimation to hypoxia, the early signs of erythropoiesis werealso assessed during IA and three paradigms of RA. Most of ourdata do not support either of these hypotheses. Rather, an argu-ment is presented that previous exposure to hypoxia can havea detrimental, rather than facilitative, impact on haematologicalre-acclimation to hypoxia.

4.1. Haematological acclimation during IA

As anticipated, the initial acclimation to hypoxia resulted ina stimulation of erythropoiesis, evidenced by a rapid increase inserum [EPO] followed by an elevated RC. As has been reported fromhuman studies (Abbrecht and Littell, 1972; Milledge and Cotes,1985), serum [EPO] did not remain elevated throughout the expo-sure but dropped back towards, and even below, BL by 14 d inhypoxia. Although RC generally follows this same pattern (Huffet al., 1975; Mylrea and Abbrecht, 1970), in IA it remained elevatedthrough 14 d hypoxia, rather than falling to BL as expected. Classicpatterns of haematological acclimation were also seen for RBC, [Hb]and Hct with a rapid initial increase in each variable followed bya more gradual rise to a plateau (see Grover and Bärtsch, 1996 forreview). MCV, RDW and MCH increased transiently, followed by agradual decrease after 1–2 w back towards BL.

Increased [EPO] following hypoxic exposure presumably waspreceded by increased EPO gene expression at the kidney (Schusteret al., 1987) and although the expected transient increase in kidneyEPO mRNA was demonstrated, a large degree of inter-individualvariability prohibited the detection of a significant effect of timein hypoxia. Given that experimental groups were age- and sex-matched cage mates from an inbred mouse strain, the degreeof within-group variability was unexpected. Variability in geneexpression between genetically identical cage mates has beenattributed to differences in immune status of individual mice(Pritchard et al., 2001) and could also reflect differences in maternalenvironment or hierarchical structures within the cage. Pritchardet al. (2001) have also reported an order effect with differentialgene expression between the first and last cage mates to be sac-rificed. Although the short time frame between sacrifices likelyprecludes an effect of time of day on gene expression, it is pos-sible that some rapid changes in gene expression occur in responseto the removal and sacrifice of cage mates. Schuster et al. (1987)reported a decrease in EPO mRNA within 1 h of the cessation ofhypoxia. In the current study, the time between removing animalsfrom the hypoxic chamber and RNA-stabilizing the excised kidneysranged from about 5–45 min but there was no observable pattern ofdecreasing EPO mRNA with animals sacrificed after longer periodsof time out of the chamber.

�-Actin is commonly used as an endogenous reference genebecause of its known stability during hypoxic exposure (Arvidssonet al., 2011; Chan et al., 2010; To and Huang, 2005). However,a recent study challenged the previously reported stability of �-actin expression in human chondrocytes during exposure to severe(FIO2 = 0.01) hypoxia (Foldager et al., 2009). Although the stabilityof �-actin in mouse peritubular cells following exposure to moremoderate hypoxia (FIO2 = 0.12) has not been explicitly refuted, theuse of an alternative, potentially more stable, reference gene such as28S rRNA (Zhong and Simons, 1999) or RNA polymerase II (Radonicet al., 2004) might have reduced inter-individual variability andimproved our ability to detect an effect of time in hypoxia on EPOmRNA expression.

Alternatively, increased EPO mRNA expression may not berequired to elicit an increase in [EPO] since Chikuma et al. (2000)

suggest that an elongated half-life of EPO mRNA may also con-tribute to increased EPO production. Furthermore, although thekidney is the predominant source of EPO gene expression and pro-tein synthesis, other tissues including the liver, brain, lung, and

ology &

rtE

4

ppFrbhth

otaAha(iItt[tacrauceo4gahinort

rrt[eCwp(tioMttTeDi

M.J. MacNutt et al. / Respiratory Physi

eproductive organs have been discovered as sites of EPO produc-ion in adult mammals and may account for up to 10% of circulatingPO (reviewed in Weidemann and Johnson, 2009).

.2. Reduced [EPO] response in RA treatments compared to IA

In all three RA treatments, [EPO] followed the same generalattern as in IA but the magnitude of the response was blunted:eak [EPO] was reduced 25% in RA (NS) and 33% in RA↓DA (p = 0.05,ig. 2). In RA↑IA, [EPO] again followed the same pattern as in IA butemained well below BL throughout the re-exposure. It is possi-le that missing data concealed an early peak in [EPO] after 1 d inypoxia (as in RA) but if the shape of the [EPO] response was consis-ent across treatments it is unlikely that this missing value wouldave approached the peak [EPO] seen in IA.

Although this is the first study to compare the early stagesf erythropoiesis throughout an initial and re-acclimation to sus-ained hypoxia, other work has compared the EPO response tocute hypoxia before and 1 w after a 3-w expedition to 6768 m.s in the present work, Savourey et al. (2004) reported that, inumans, [EPO] was reduced following DA and the EPO response tocute re-exposure was diminished. In our previous work in humansMacNutt et al., 2009, 2012), a reduction in [EPO] also occurred dur-ng the early stages of DA but was restored after 10 d at low altitude.n this case, the [EPO] response during an extended re-exposureo high altitude was also slightly blunted compared to IA. Unlikehe present work, Savourey et al. (2004) also reported that RBC,Hb] and Hct had not fully de-acclimatised before re-exposure andhat PaCO2 was lower and PaO2 , SaO2 and CaO2 were higher duringcute hypoxia after the expedition as compared to before. They con-luded that individuals were still partially acclimatised when weree-exposed to hypoxia, and better-protected arterial oxygenationnd higher CaO2 at the kidney resulted in a lesser hypoxic stim-lus and a blunted EPO response. This might also have been thease in the previous field work in humans, since [Hb] was slightlylevated upon re-introduction to high altitude. The same was truef mine workers exposed to intermittent sustained hypoxia (7 d at200 m, 7 d at SL) for 2.5 y (Richalet et al., 2002). [Hb] and Hct pro-ressively increased at SL and altitude while the EPO response toltitude re-exposure tended to fall over time. In the present study,aematological variables had returned to BL before re-exposure but

t is possible that improvements in aspects of acclimation that wereot measured here (i.e. minute ventilation, blood flow distribution,xyhemoglobin affinity) also resulted in higher CaO2 and therefore aeduced erythropoietic stimulus at the kidney during RA comparedo IA.

However, Savourey et al. (2004) still reported a reduced [EPO]esponse for a given CaO2 , and therefore a true blunting of the [EPO]esponse reflecting decreased sensitivity of EPO-producing cellso hypoxia. This could also account for the common finding thatEPO] drops towards or below BL after 1–2 w of sustained hypoxicxposure (Abbrecht and Littell, 1972; Milledge and Cotes, 1985).onversely, several studies have reported that the [EPO] responseas considerably less blunted in mice that were made similarlyolycythemic by recent hypoxic exposure compared to transfusionreviewed in Bozzini et al., 1994), suggesting an increased sensi-ivity of the EPO response for a given hypoxic stimulus. Althoughn opposition to findings by Savourey and colleagues and to ourwn previous and current work, these results a hard to refute;artinez et al. (2010) demonstrated a clear dose–response rela-

ionship between severity of the previous hypoxic exposure andhe increased sensitivity of the [EPO] response to re-exposure.

he discrepancy in findings may relate to the timing of re-xposure. A blunted [EPO] response has been reported after 7–14 dA (MacNutt et al., 2009, 2012; Savourey et al., 2004) whereas

ncreased sensitivity was demonstrated after only 24–72 h DA

Neurobiology 189 (2013) 153– 161 159

(Alippi et al., 1983; Martinez et al., 2010). In the present study, the[EPO] response was most blunted in RA↑IA suggesting the degreeto which sensitivity is altered also depends upon the duration ofIA. Clearly the effects of previous hypoxic exposure on subsequentsensitivity of EPO-producing organs require further elucidation.Furthermore, the implications of altered [EPO] responses duringRA for a number of increasingly recognised non-erythropoieticfunctions of EPO (reviewed in Vogel and Gassmann, 2011) meritinvestigation. Our sample size was based on an a priori sample sizecalculation to produce 80% statistical power to detect the expectedchanges in [Hb] (means and standard deviations estimated fromour human field study, MacNutt et al., 2012). We acknowledge thatthe absence of difference between conditions likely reflects a higherthan anticipated degree of between-animal variation.

4.3. Similar reticulocyte response despite blunted [EPO] in RAtreatments compared to IA

Despite the blunted [EPO] response during re-exposure, RCpeaked at the same value (7–8%) in RA and RA↑IA as in IA. Unfor-tunately, data for RA↓DA were unavailable for the 7 d time pointso it is uncertain whether the reticulocyte response for this treat-ment was consistent with the other groups. Although their timeframe was much shorter, Savourey et al. (2004) actually reportedan increased reticulocyte response in spite of reduced [EPO] duringre-exposure. One explanation offered was that increased sensi-tivity of EPO receptors on erythroid progenitor cells resulted inenhanced differentiation for a given EPO stimulus (Savourey et al.,2004). In support, Okunewick and Fulton (1970) demonstrated thatthe erythropoietic response to exogenous EPO (measured by 59Feuptake) was greater in mice recently made similarly polycythemicby hypoxic exposure versus transfusion. The current data set pro-vides evidence that erythropoiesis might be regulated by relative,rather than absolute changes in [EPO]. Despite large differences inabsolute values, the relative increases in [EPO] from pre-exposureto peak values were remarkably similar between treatment groupsthat exhibited a consistent reticulocyte response (130, 140, and140% increase for IA, RA and RA↑IA, respectively). These data suggestthat EPO receptors were sensitised to reduced circulating levels of[EPO] during DA and responded to the relative increase in [EPO] thatoccurred in each RA. Soluble EPO receptors (sEPOR) compete withEPO receptors (EPOR) to bind EPO. It appears that sEPOR down-regulation in the brain is required for ventilator acclimatisation tohypoxia in mice (Soliz, 2013). We did not assess sEPOR in our studybut it is possible that our results also reflect a changing level ofsEOPR.

4.4. Decreased [Hb] and MCH throughout RA↑IA compared to IA

As shown in IA and in previous studies (Marki et al., 1982;Schobersberger et al., 2005), MCH showed a transient increase inthe early stages of RA and RA↓DA before decreasing back to BL(Fig. 5). In contrast, the [Hb] response was significantly reducedby ∼9% across time points in RA↑IA compared to IA with a per-sistent drop in MCH below BL, despite a similar RBC response.Thus, contrary to our original hypothesis, O2 carrying capacity wasreduced, rather than increased in RA↑IA compared to IA. This is notthe first report that previous hypoxic exposure can have apparentlyadverse, rather than beneficial, effects on re-acclimation responses(Prabhakar and Semenza, 2012). Here we emphasise that the phys-iological responses to a hypoxic stimulus depend upon the type ofhypoxic exposure. Specifically, hypoxia can be acute or chronic and

continuous or intermittent and these must be considered. Our studyemployed a model of intermittent hypoxia which was comparableto that of others. Using a model of intermittent hypoxic expo-sure in rabbits, Jain et al. (1978) observed a faster haematological

1 ology &

reRwitpafamfrabat

nrttssarltttbeaiiht

4I

bHRbiaairwRamaMMr

5

wa

60 M.J. MacNutt et al. / Respiratory Physi

esponse with a larger increase in [Hb] in RA compared to IA. How-ver, there was no evidence of increased erythropoiesis duringA and the increased [Hb] response was driven by four animalsho exhibited a stronger than average hemoconcentration dur-

ng RA. In fact, hypovolemia in those individuals was so severehat they died during RA, apparently from hyperviscosity-inducedulmonary haemorrhage. An increased hemoconcentration duringcute hypoxic re-exposure has also been demonstrated in humansollowing continuous initial exposure to HA; however, this wasssociated with the positive outcome of reducing severity of acuteountain sickness during RA (Lyons et al., 1995). Conversely, data

rom Savourey et al. (2004) suggest a reduced hemoconcentrationesponse during acute re-exposure following a 3-w sojourn to highltitude. We report no measures of PV or fluid balance; however,ased on the repeatability of early changes in RBC, Hct and [Hb]cross treatments, there is no evidence of altered hemoconcentra-ion during RA groups compared to IA.

The reasons for decreased [Hb] and MCH throughout RA↑IA areot entirely clear but a transient decrease in MCH has also beeneported in the pig-tailed monkey (Buderer and Pace, 1972) and inhe mouse (Mylrea and Abbrecht, 1970). Interestingly, in the lat-er study this pattern only occurred in animals exposed to moreevere hypoxia (∼6200 m) and not during milder hypoxic expo-ures (∼4600 and ∼3400 m). The greater hypoxic stimulus elicited

larger reticulocyte response, including relatively more “shift”eticulocytes than following less hypoxic exposures. Shift reticu-ocytes are prematurely released from the bone marrow duringimes of particular erythropoietic stress (Turgeon, 1999) and con-ain less Hb than normal reticulocytes (Skadberg et al., 2003). Ashese shift reticulocytes mature they continue to synthesise Hb,ringing MCH back towards BL. Given the identical environmentalxposure, the reduced [EPO] response, and the similar reticulocytend RBC response between RA↑IA and the other treatments, theres no evidence to suggest that hypoxic stimulus was more severen RA↑IA. However, the extended initial exposure to hypoxia mayave depleted the available pool of mature reticulocytes, leading tohe premature release of more shift reticulocytes during RA↑IA.

.5. Reduced MCV and Hct but increased RDW in RA compared toA

The response patterns of MCV, Hct and RDW were similaretween IA and RA but the magnitude of responses differed.ypoxia-induced increases in MCV were significantly reduced inA compared to IA and, since the RBC response was the same foroth treatments, the outcome was a slightly reduced Hct response

n RA (Fig. 6). The transient increases in both MCV and RDW thatre reported here and elsewhere (Marki et al., 1982), could bettributable to the hypoxia-induced reticulocytosis that occurs dur-ng the early stages of hypoxic exposure; the presence of largeeticulocytes elevates MCV and increases the variability of cell sizeithin the red cell population (Turgeon, 1999). The newly releasedBCs get smaller as they mature, leading to a decrease in both MCVnd RDW. However, reported increases in MCV of 4–10 fL cannot beathematically explained by the observed reticulocytosis alone. In

ddition, other studies in mice (Huff et al., 1975; McLaughlin andeints, 1972; Mylrea and Abbrecht, 1970) report that the increasedCV persists throughout the hypoxic exposure, even after RC has

eturned to normal (Mylrea and Abbrecht, 1970).

. Conclusions

We hypothesised that haematological acclimation to hypoxiaould be facilitated by recent exposure to the same stimulus. Using

similar time frame as our recent human field study (MacNutt et al.,

Neurobiology 189 (2013) 153– 161

2012), the finding of increased [Hb] in RA compared to IA couldnot be confirmed in controlled laboratory experiments with mice.Nor could the hypothesis that decreasing the duration of DA and/orincreasing the duration of IA would further improve haematologicalacclimation during RA be confirmed. Instead, evidence is presentedthat haematological re-acclimation can actually be impaired rela-tive to an initial acclimation, as demonstrated by a reduction in [Hb]and MCH throughout RA↑IA compared to IA. This corresponded witha normal reticulocyte and RBC response, suggesting that impairedHb synthesis in early erythrocytes was responsible.

Acknowledgements

This work was supported by the Natural Sciences and Engineer-ing Research Council of Canada. The authors wish to thank thestaff at the UBC Zoology Animal Care Centre. Many thanks to AndyJohnson for technical assistance with flow cytometry and to Drs.Trish Schulte and Chantal Leger for valuable discussions about datainterpretation.

References

Abbrecht, P.H., Littell, J.K., 1972. Plasma erythropoietin in men and mice duringacclimatization to different altitudes. Journal of Applied Physiology 32, 54–58.

Alippi, R.M., Barcelo, A.C., Bozzini, C.E., 1983. Enhanced erythropoiesis induced byhypoxia in hypertransfused, post-hypoxic mice. Experimental Hematology 11,878–883.

Arvidsson, Y., Bergstrom, A., Arvidsson, L., Kristiansson, E., Ahlman, H., Nilsson,O., 2011. Hypoxia stimulates CXCR4 signalling in ileal carcinoids. Endocrine-Related Cancer 17, 303–316.

Bozzini, C.E., Alippi, R.M., Barcelo, A.C., Conti, M.I., Bozzini, C., Lezon, C.E., Olivera,M.I., 1994. The biology of stress erythropoiesis and erythropoietin production.Annals of the New York Academy of Sciences 718, 83–92, discussion 92-83.

Buderer, M.C., Pace, N., 1972. Hemopoiesis in the pig-tailed monkey Macaca nemest-rina during chronic altitude exposure. American Journal of Physiology 223,346–352.

Bult, C.J., Eppig, J.T., Kadin, J.A., Richardson, J.E., Blake, J.A., 2008. The Mouse GenomeDatabase (MGD): mouse biology and model systems. Nucleic Acids Research 36,D724–D728.

Chan, S., Ngo, T.T.T., Cameron, J.S., 2010. Altered gene expression in the heart afteracclimation to hypoxia in goldfish: evaluation of candidate reference genes forqPCR studies. FASEB Journal 24 (Abstract), 988.912.

Chikuma, M., Masuda, S., Kobayashi, T., Nagao, M., Sasaki, R., 2000. Tissue-specificregulation of erythropoietin production in the murine kidney, brain, anduterus. American Journal of Physiology – Endocrinology and Metabolism 279,E1242–E1248.

Foldager, C.B., Munir, S., Ulrik-Vinther, M., Soballe, K., Bunger, C., Lind, M., 2009.Validation of suitable house keeping genes for hypoxia-cultured human chon-drocytes. BMC Molecular Biology 10, 94.

Grover, R.F., Bärtsch, P., 1996. Blood. In: Fregly, M.J., Blatteis, C.M. (Eds.), Hand-book of Physiology. Section 4. Environmental Physiology, vol. 1, 2nd ed. OxfordUniversity Press, New York, pp. 493–523.

Hannon, J.P., Chinn, K.S., Shields, J.L., 1969. Effects of acute high-altitude exposureon body fluids. Federation Proceedings 28, 1178–1184.

Hedrich, H.J., 2004. The laboratory mouse. In: Bullock, G.R., Petrusz, P. (Eds.), TheHandbook of Experimental Animals. Elsevier Academic Press, London, UK, p.600.

Hoyt, R., Honig, A., 1996. Body fluid and energy metabolism at high altitude. In:Fregly, M.J., Blatties, C.M. (Eds.), Handbook of Physiology. Section 4. Environ-mental Physiology, vol. 2. Oxford University Press, New York, pp. 1277–1289.

Huff, J.E., Kaufman, G.E., Ingram, M., 1975. Hematologic changes in mice during andafter exposure to severe hypobaric hypoxia. Aviation Space and EnvironmentalMedicine 46, 1147–1151.

Jain, S.C., Malhotra, M.S., Krishna, B., Bardhan, J., Grover, A., 1978. Hematologicchanges in rabbits during acclimatization, deacclimatization, and reinductionto hypoxia. Aviation Space and Environmental Medicine 49, 952–955.

Jelkmann, W., 2011. Regulation of erythropoietin production. Journal of Physiology589, 1251–1258.

Knaupp, W., Khilnani, S., Sherwood, J., Scharf, S., Steinberg, H., 1992. Erythropoietinresponse to acute normobaric hypoxia in humans. Journal of Applied Physiology73, 837–840.

Koury, S.T., Bondurant, M.C., Semenza, G.L., Koury, M.J., 1993. The use of in situhybridization to study erythropoietin gene expression in murine kidney and

liver. Microscopy Research and Technique 25, 29–39.

Lacombe, C., Da Silva, J.L., Bruneval, P., Fournier, J.G., Wendling, F., Casadevall, N.,Camilleri, J.P., Bariety, J., Varet, B., Tambourin, P., 1988. Peritubular cells are thesite of erythropoietin synthesis in the murine hypoxic kidney. Journal of ClinicalInvestigation 81, 620–623.

ology &

L

L

M

M

M

M

M

M

M

O

P

P

R

R

R

M.J. MacNutt et al. / Respiratory Physi

i, X., Kimura, H., Hirota, K., Kasuno, K., Torii, K., Okada, T., Kurooka, H., Yokota, Y.,Yoshida, H., 2005. Synergistic effect of hypoxia and TNF-alpha on production ofPAI-1 in human proximal renal tubular cells. Kidney International 68, 569–583.

yons, T.P., Muza, S.R., Rock, P.B., Cymerman, A., 1995. The effect of altitude pre-acclimatization on acute mountain sickness during reexposure. Aviation Spaceand Environmental Medicine 66, 957–962.

acNutt, M.J., Laursen, P.B., Kedia, S., Neupane, M., Parajuli, P., Pokharel, J., Sheel,A.W., 2012. Acclimatisation in trekkers with and without recent exposure tohigh altitude. European Journal of Applied Physiology 112, 3287–3294.

acNutt, M.J., Laursen, P.B., Kedia, S., Neupane, M., Pokharel, J., Parajuli, P., Sheel,A.W., 2009. Destination Gokyo: acclimatisation and re-acclimatisation in highaltitude trekkers. In: International Hypoxia Symposium, Lake Louise, Alberta(Abstract).

arki, U., Weihe, W.H., Schneider, J., Stranzinger, G., 1982. The haematological andcardiac patterns of adaptation of rabbits to 4,500 m simulated altitude and ofreadaptation to 400m. Research in Experimental Medicine 181, 197–204.

artinez, M.P., Conti, M., Barcelo, A.C., Bozzini, C.E., 2010. Hypoxia-stimulatederythropoietin secretion in mice with different types of polycythemia: theposthypoxic enigma. Advanced Studies in Biology 2 (1), 35–52.

cLaughlin, D.W., Meints, R.H., 1972. A study of hibernator erythropoietic responsesto simulated high altitude. Comparative Biochemistry and Physiology Part A:Physiology 42, 655–666.

illedge, J.S., Cotes, P.M., 1985. Serum erythropoietin in humans at high altitudeand its relation to plasma renin. Journal of Applied Physiology 59, 360–364.

ylrea, K.C., Abbrecht, P.H., 1970. Hematologic responses of mice subjected to con-tinuous hypoxia. American Journal of Physiology 218, 1145–1149.

kunewick, J.P., Fulton, D., 1970. Comparison of erythropoietin response in micefollowing polycythemia induced by transfusion of hypoxia. Blood 36, 239–245.

rabhakar, N.R., Semenza, G.L., 2012. Adaptive and maladaptive cardiorespiratoryresponses to continuous and intermittent hypoxia mediated by hypoxia-inducible factors 1 and 2. Physiological Reviews 92, 967–1003.

ritchard, C.C., Hsu, L., Delrow, J., Nelson, P.S., 2001. Project normal: defining normalvariance in mouse gene expression. Proceedings of the National Academy ofSciences of the United States of America 98, 13266–13271.

adonic, A., Thulke, S., Mackay, I.M., Landt, O., Siegert, W., Nitsche, A., 2004. Guide-line to reference gene selection for quantitative real-time PCR. Biochemical andBiophysical Research Communications 313, 856–862.

ice, L., Ruiz, W., Driscoll, T., Whitley, C.E., Tapia, R., Hachey, D.L., Gonzales, G.F.,

Alfrey, C.P., 2001. Neocytolysis on descent from altitude: a newly recognizedmechanism for the control of red cell mass. Annals of Internal Medicine 134,652–656.

ichalet, J.P., Vargas, M., Cortes, G., Leon, T., Hudson, C., Jimenez, D., 2002. Hema-tological response in Chilean mine workers in intermittent hypoxia (IH).

Neurobiology 189 (2013) 153– 161 161

In: World Congress on Mountain Medicine and High Altitude Physiology,Barcelona.

Savourey, G., Garcia, N., Besnard, Y., Guinet, A., Hanniquet, A.M., Bittel, J., 1996. Pre-adaptation, adaptation and de-adaptation to high altitude in humans: cardio-ventilatory and haematological changes. European Journal of Applied Physiologyand Occupational Physiology 73, 529–535.

Savourey, G., Launay, J.C., Besnard, Y., Guinet, A., Bourrilhon, C., Cabane, D., Martin, S.,Caravel, J.P., Pequignot, J.M., Cottet-Emard, J.M., 2004. Control of erythropoiesisafter high altitude acclimatization. European Journal of Applied Physiology 93,47–56.

Schobersberger, W., Greie, S., Humpeler, E., Mittermayr, M., Fries, D., Schobers-berger, B., Artner-Dworzak, E., Hasibeder, W., Klingler, A., Gunga, H.C., 2005.Austrian Moderate Altitude Study (AMAS 2000): erythropoietic activity and Hb-O2 affinity during a 3-week hiking holiday at moderate altitude in persons withmetabolic syndrome. High Altitude Medicine and Biology 6, 167–177.

Schuster, S.J., Wilson, J.H., Erslev, A.J., Caro, J., 1987. Physiologic regulation and tissuelocalization of renal erythropoietin messenger RNA. Blood 70, 316–318.

Sheel, A.W., MacNutt, M.J., 2008. Control of ventilation in humans following inter-mittent hypoxia. Applied Physiology, Nutrition, and Metabolism 33, 573–581.

Skadberg, O., Brun, A., Sandberg, S., 2003. Human reticulocytes isolated from periph-eral blood: maturation time and hemoglobin synthesis. Laboratory Hematology9, 198–206.

Soliz, J., 2013. Erythropoietin and respiratory control at adulthood and during earlypostnatal life. Respiratory Physiology and Neurobiology 185, 87–93.

To, K.K., Huang, L.E., 2005. Suppression of hypoxia-inducible factor 1alpha (HIF-1alpha) transcriptional activity by the HIF prolyl hydroxylase EGLN1. Journal ofBiological Chemistry 280, 38102–38107.

Turgeon, M.L., 1999. Clinical Hematology: Theory and Procedures, third ed. Lippin-cott Williams & Wilkins, New York.

Van Hove, L., Goossens, W., Van Duppen, V., Verwilghen, R.L., 1990. Reticulocytecount using thiazole orange. A flow cytometry method. Clinical and LaboratoryHaematology 12, 287–299.

Vogel, J., Gassmann, M., 2011. Erythropoietic and non-erythropoietic functions oferythropoietin in mouse models. Journal of Physiology 589, 1259–1264.

Ward, N.L., Moore, E., Noon, K., Spassil, N., Keenan, E., Ivanco, T.L., LaManna, J.C.,2007. Cerebral angiogenic factors, angiogenesis, and physiological response tochronic hypoxia differ among four commonly used mouse strains. Journal ofApplied Physiology 102, 1927–1935.

Weidemann, A., Johnson, R.S., 2009. Nonrenal regulation of EPO synthesis. KidneyInternational 75 (7), 682–688.

Zhong, H., Simons, J.W., 1999. Direct comparison of GAPDH, beta-actin, cyclophilin,and 28S rRNA as internal standards for quantifying RNA levels under hypoxia.Biochemical and Biophysical Research Communications 259, 523–526.