FGF-21 and skeletal remodeling during and after lactation in C57BL/6J mice

FGF-21 and Skeletal Remodeling During and AfterLactation in C57BL6 Mice

Sheila Bornstein, MS,1, Sue A. Brown, MD,2, Phuong T. Le, MS,1,Xunde Wang, PhD,3, Victoria DeMambro, MS,1, Mark C. Horowitz, PhD,4,Ormond MacDougald, PhD,5, Roland Baron, DDS, PhD,6, Sutada Lotinun, PhD,6,Gerard Karsenty, MD, PhD,7, Wei Wei, PhD3, Mathieu Ferron, PhD7,8,9,Christopher S. Kovacs, MD,10, David Clemmons, MD,11, Yihong Wan, PhD3*,and Clifford J. Rosen, MD1*1Maine Medical Center Research Institute, Scarborough, ME; 2University of Virginia Division ofEndocrinology, Charlottesville, VA; 3University of Texas Southwestern, Division of Pharmacology, Dallas,TX; 4Yale University School of Medicine, Department of Orthopaedics and Rehabilitation, New Haven,CT; 5University of Michigan Metabolism, Endocrinology and Diabetes Division, Ann Arbor, MI; 6HarvardDental School, Boston, MA; 7Columbia University College of Physicians and Surgeons, Department ofGenetics, NY, NY; 10Memorial University, Division of Endocrinology, St. John’s, Newfoundland CanadaA1B 3V6; 11University of N Carolina Chapel Hill, Division of Endocrinology, Chapel Hill, NCKey words: bone, lactation, FGF21

Lactation is associated with significant alterations in both body composition and bone mass. Sys-temic and local skeletal factors such as RANKL, PTHrP, calcitonin, and estrogen are known toregulate bone remodeling during and after lactation. FGF21 may function as an endocrine factorto regulate body composition changes during lactation by inducing gluconeogenesis and fatty acidoxidation. In this study, we hypothesized that the metabolic changes during lactation were due inpart to increased circulating FGF-21, which in turn could accentuate bone loss. We longitudinallycharacterized body composition in C57BL/6J (B6) mice during (d7Lac and d21Lac) and after normallactation (d21Post). At day 7(d7Lac) of lactation, areal bone density declined by 10% (aBMD:p�0.001) bone resorption increased (p�0.0001), percent fat decreased by 20%, energy expendi-ture increased (p�0.01), and markers of brown-like adipogenesis were suppressed in the inguinaldepot and in pre-formed brown adipose tissue. At d7Lac there was a 2.4 fold increase in serumFGF-21 vs. baseline (p�0.0001), a 8-fold increase in hepatic FGF-21 mRNA (p�0.03), a 2-fold increasein undercarboxylated osteocalcin (p�0.01) and enhanced insulin sensitivity. Recovery of totalaBMD was noted at d21Lac, while femoral trabecular bone volume fraction (BV/TV) was still re-duced (p�0.01). Since FGF-21 levels rose rapidly at d7Lac in B6 lactating mice, we next examinedlactating mice with a deletion in the Fgf21 gene. Trabecular as well as cortical bone mass weremaintained throughout lactation in FGF-21-/- mice, while pup growth was normal. Compared tolactating control mice, lactating FGF-21-/- mice exhibited an increase in bone formation, but nochange in bone resorption. In conclusion, in addition to changes in calcitropic hormones, systemicFGF-21 plays a role in skeletal remodeling and changes in body composition during lactation in B6mice.

Lactation is essential for mammalian survival (1, 2). Pre-vious work established there was uncoupling of bone

remodeling during the early phases of lactation in mice,

rats and humans, with high rates of bone resorption anddecreased bone formation (3–7). Uncoupling in skeletalremodeling has been variously ascribed to high circulating

ISSN Print 0013-7227 ISSN Online 1945-7170Printed in U.S.A.Copyright © 2014 by the Endocrine SocietyReceived January 30, 2014. Accepted June 3, 2014.

Abbreviations:

G E N E R A L E N D O C R I N O L O G Y

doi: 10.1210/en.2014-1083 Endocrinology endo.endojournals.org 1

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 10 June 2014. at 14:15 For personal use only. No other uses without permission. . All rights reserved.

RANKL concentrations, low estrogen levels, and in-creased PTHrP, although other factors are likely to beoperative (8–10). The postlactation period is character-ized by skeletal recovery through poorly understoodmechanisms although it has been related to osteoclast ap-optosis, refilling of lacunae by osteocytes, and an inde-pendent anabolic-rebuilding period (4, 11–12). Severalcandidates have been suggested as mediators of the latterstate including estrogen, oxytocin, PTHrP, IGF-I, and oth-ers (11, 13, 14). Conditional deletion of PTHrP in mam-mary tissue reduced the magnitude of bone loss betweendays 4–12 of lactation but the extent of skeletal recoverywas not examined (8, 15). It is also unclear what factorsregulate the body composition changes that accompanylactation. FGF21 is a member of a subfamily of FGFs(which include FGF15/19 and FGF23), which lack theFGF heparin-binding domain. FGF21 functions as an en-docrine factor by inducing gluconeogenesis and fatty acidoxidation during stressful conditions such as fasting,when increased substrate utilization is required to main-tain homeostasis (16). In this study, we hypothesized thatthe metabolic changes during lactation were due in part toincreased circulating FGF-21, which in turn could accen-tuate bone loss.

Immediately after lactation there are also dramaticchanges in body composition and a shift in energy utili-zation from mammary tissue towards bone to ensure skel-etal restoration. In this study we systematically examinedwhether the skeletal changes during lactation could berelated to metabolic determinants as well as calciotropicfactors. Our results confirm there are profound, but re-versible changes in whole body and skeletal metabolismduring and after lactation. We also found that FGF-21 isone of several factors modulating substrate utilization andbone turnover during lactation.

Materials and Methods

AnimalsC57BL/6J (B6) mice were purchased from The Jackson Lab-

oratory (Bar Harbor, ME). Eight-week-old C57BL/6J femalemice were mated and after the mice became pregnant, males wereremoved. After birth lactating females were studied at day 7(d7Lac), day 21 (d21Lac) and day 21-post lactation (d21Post).Mice were weaned at 21 days after birth, therefore the d7Lactime point examines early lactation and the d21Lac time pointexamines late lactation, assuming lactation ends shortly afterweaning at 21 days after birth. The d21Post time point examinesthe effect of lactation 21 days post wean. Age-matched nonlac-tating C57BL/6J control mice were also maintained for each timepoint (d7Lac vs. 12wk nonlactating control; d21Lac vs. 15wknonlactating control and d21Post vs. 20wk nonlactating con-trol). All age matched animals were � 5 days of lactating ani-

mals, this allows us to take into consideration variation of thetime of conception. FGF-21 null mice have been described (16).The FGF-21 null mice were kept on a B6 background (back-crossed for at least 6 generations). FGF-21 KO and WT litter-mates from the last backcross were then bred as WTxWT andKOxKO, which provided the mice for all the FGF-21 studies.FGF-21 KO and WT mice were set up for mating at 10 weeks ofage and samples were harvested at day 14 of lactation (d14Lac)with age-matched nonlactating virgin controls (Ctrl; 16wks old).Mice were maintained at 25°C on a 14-hour dark, 10-hour lightcycle on standard regular chow (Harlan Tekland, Indianapolis,IN, USA). All the mice were fed standard chow containing 4% fatad libitum. All experimental studies and procedures involvingmice were reviewed and approved by the Institutional AnimalCare and Use Committee of MMCRI (Scarborough, ME, USA)and The University of Texas Southwestern Medical Center (Dal-las, TX, USA).

Dual-energy X-ray absorptiometryDual-energy X-ray absorptiometry (DXA) for whole-body

and femoral areal bone mineral density (BMD) (aBMD, mg/cm2)and body composition exclusive of the head were performedusing the PIXImus (GE Lunar, Fairfield, CT, USA) as describedpreviously (10). The PIXImus was calibrated daily with a phan-tom provided by the manufacturer.

Microcomputed tomography (�CT)Microarchitecture of the distal trabecular bone and midshaft

cortical bone of the femur was analyzed by �CT (resolution 10�m, VivaCT-40; Scanco Medical AG, Bassersdorf, Switzerland)as published previously (17). The VivaCT-40 is calibratedweekly using a phantom provided by Scanco Medical AG. Tra-becular BV/TV and microarchitecture were evaluated in the sec-ondary spongiosa, starting proximately at 0.6 mm proximal tothe distal femoral growth plate, and extending proximally 1.5mm. Measurements included bone volume/total volume (BV/TV), trabecular number (Tb.N.), trabecular thickness (Tb.Th.)and trabecular separation (Tb.Sp.). Scans for the cortical regionwere measured at the midpoint of each femur measuring theaverage bone area (BA), total cross-sectional area (TA), bonearea/total area (BA/TA), cortical thickness (Ct.Th.) and corticalporosity (Ct.Po.).

Osmium �CT analysis for marrow adipose tissueStaining for marrow fat requires the use of osmium tetraoxide

(OsO4) in a four-step procedure as reported previously as a mod-ification of the �CT protocol (17, 18). Mouse tibia were isolatedand then fixed in 10% neutral buffered formalin. Tibias werethen decalcified with EDTA at 4°C for 14 days. The decalcifiedbones were immersed in OsO4 for 48 hours at room temperatureprior to imaging with a VivaCT-40 (Scanco Medical AG,Bassersdorf, Switzerland) using a voxel size of 10 �m. Briefly,four volumes of interest (VOI) within the marrow space, extend-ing 230 mm, were defined by their position along the z-axis. Thefirst VOI began 230 mm from the proximal end of the tibia, andeach subsequent VOI began 1.2 mm distal to the previous VOI.The fractional volume of osmium-stained fat within the VOI’swas determined using the Scanco 3D bone morphometry eval-uation program with a threshold of 420 and Gaussian filtering(s � 0.8, support � 1).

2 Endocrinology


Bone histomorphometryStatic and dynamic histomorphometry measures were ana-

lyzed between lactating and age-matched nonlactating controlC57BL/6J mice at 12wk vs. d7Lac, 15wk vs. d21Lac and 20wkvs. d21Post lactation. FGF-21 KO and WT mice were analyzedat d14Lac vs. 13wk nonlactating controls. Mice were injectedwith 20 mg/kg calcein and demeclocycline IP 9 days and 2 days,respectively, before sample collection. Femurs were analyzed asdescribed and standard nomenclature was used (17, 19).

RNA isolation and quantitative real-time PCRTotal RNA was prepared using an RNeasy Mini Kit (Qiagen,

Valencia, CA, USA) for liver and mammary tissue or using astandard TRIzol (Sigma, St. Louis, MO, USA) method for tibias.Total RNA was extracted from lipid depots using an RNeasyLipid Tissue Mini Kit (Qiagen, Valencia, CA, USA). cDNA wasgenerated using a random hexamer and reverse transcriptase(SuperscriptIII; Invitrogen, Carlsbad, CA, USA) according to themanufacturer’s instructions. mRNA expression analysis wascarried out using an iQ SYBR Green Supermix with an iQ5 ther-mal cycler and detection system (Bio-Rad, Hercules, CA, USA).Relative expression of mRNA was determined after normaliza-tion to hypoxanthine guanine phosphoribosyl transferase (Hprt)levels in addition to �-2 microglobulin (B2M) (inguinial fat,liver), 18s (interscapular fat) or ribosomal protein L13A(Rpl13a) (mammary, tibia). Primers were designed and tested tobe 95% to 100% efficient by PrimerDesign (Southampton, UK)unless otherwise noted. All primer sequences are listed (Table SI).

Serum biochemical parametersSerum concentrations of amino-terminal propeptide of type 1

procollagen (P1NP), cross-linked C-telopeptide (CTX) and in-sulin growth factor 1 (IGF1) were measured with the Rat/MouseP1NP enzyme immunoassay (EIA), the RatLaps EIA and theMouse IGF1 High Sensitivity immunoassay (EIA), respectively(Immunodiagnostic Systems, Scottsdale, AZ, USA). The assaysensitivities were 0.7, 2 and 63 ng/ml for P1NP, CTX and IGF1respectively. The intraassay variations were 6.3%, 6.9% and6.8% and the interassay variations were 8.5%, 12% and 7.3%,respectively. Serum concentrations of insulin and sclerostin weremeasured with the Mouse Insulin immunoassay (EIA), andMouse Sclerostin immunoassay (EIA), respectively (ALPCO, Sa-lem, NH, USA). The assay sensitivities were 0.12 ng/mL and 17.4pg/mL for insulin and sclersotin, respectively. The intraassayvariations were 3.7% and 6.4% and the interassay variationswere 3.2% and 10.3% respectively. Serum concentration of fi-broblast growth factor 21 (FGF-21) was measured with theMouse/Rat FGF-21 immunoassay (EIA) (BioVendor, Candler,NC, USA). The assay sensitivity was 18.4 pg/mL, the intraassayvariation was 8.4% and interassay variation was 8.7%. Serumconcentration of osteocalcin (OCN) was measured with MouseOsteocalcin immunoassay (EIA) (Biomedical Technologies,Stoughton, MA, USA). The assay sensitivity was 1 ng/mL, theintraassay variation was 6% and the interassay variation was8%. Undercarboxylated osteocalcin (Glu13 OCn) was per-formed as previously described (20). All measurements were per-formed in duplicate.

Indirect calorimetryIndirect calorimetry measurements were performed using the

Promethion Metabolic Cage System (Sable Systems, Las Vegas,NV, USA) located in the Physiology Core Department of MaineMedical Research Institute as previously described (21). Micewere subjected to a standard 14h-light/10h-dark cycle during thestudy, which consisted of a 12-hour acclimation period followedby a 72-hour sampling duration. Data shown are representativeof the 24-hour average of this period. Correlations were per-formed between energy expenditure (EE) and lean, fat and totalbody mass to determine if these variables could explain changesin energy expenditure (22–23)

Insulin tolerance testsC57BL/6J mice were tested for insulin tolerance at d7Lac and

compared to age-matched 12wk old nonlactating controls. Micewere fed ad libitum and injected I.P. with insulin at a dose of1U/Kg. Glucose was measured at 0, 15, 30, 45, 60, and 120minutes post injection using an OneTouch Ultra Glucometer(LifeScan, Inc., Milpitas, California) (24).

Statistical AnalysisData are expressed as mean � S.E.M. Results were analyzed

for statistically significant differences using Student’s t test orANOVA followed by Tukey’s multiple comparison test orKruskal-Wallis test with Dunn’s multiple comparison test whereappropriate. Multiple regression analysis was used to adjust forgroup and lean mass for energy expenditure studies. All statistics,including regression analysis, were performed with Prism 6 sta-tistical software (Graph Pad Software, Inc., La Jolla, CA, USA).Statistical significance was set at P � .05.

Results

Areal BMD was lost and then regained in lactatingmice

For areal bone density measurements by DXA, B6 mice(n � 9–11/group) were measured longitudinally at bi-weekly intervals during lactation and post lactation. Thecontrols were nonlactating B6 mice of the same age andgender measured longitudinally. Total areal BMD weresignificantly reduced at d7Lac (P � .001) compared tononlactating controls at 12 weeks of age and by d21Lachad returned to d7Lac levels (Table 1). At d21Lac, thedifference in aBMD (femoral and total) compared to non-lactating controls at 15 weeks persisted (Table 1; P �.001). During the weaning period at d21Post, total andfemoral aBMD increased significantly compared to eitherd7Lac or d21Lac measurements, but femoral aBMD didnot reach levels in the 20wk nonlactating controls (Table1).

Lactation alters microarchitecture throughtrabecular and cortical bone loss

B6 mice were sacrificed at specific time intervals duringand post lactation that were identical to the longitudinal

doi: 10.1210/en.2014-1083 endo.endojournals.org 3


studies by DXA. Hence the �CT and histomorphometrywere cross-sectional studies of cohorts at distinct ages.Femoral BV/TV was lower in the lactating mice vs. con-trols at all time points (Table 2, Figure 1A; P � .05).Consistent with the changes in trabecular bone volumefraction, lactatingmicehad lower connectivitydensity andlower trabecular number, at d7Lac compared to controls(Table 2, P � .05). At d7Lac mice also displayed a non-significant trend for higher trabecular spacing and thinnertrabeculae compared to controls (Table 2). TrabecularBV/TV and trabecular number continued to decrease, andtrabecular spacing continued to increase, at both d21Lacand d21Post (compared to d7Lac and age-matched con-trols). In the femoral midshaft, lactating mice had a non-significant increase in total cross sectional area (Tt.Ar,Table 2) but reduced cortical thickness and cortical areafraction at d7Lac (Table 2). At d21Lac, cortical thicknesswas reduced further (P � .005 compared to d7Lac) and

remained significantly lower than controls. However, atd21Post, cortical thickness was significantly improvedcompared to d21Lac (Table 2). Cortical porosity was sig-nificantly higher in lactating animals compared to con-trols, however by d21Post lactation porosity was reducedcompared to d21Lac, (Ct.Po, Table 2). At d21Post, tra-becular bone mass and microarchitecture did not fully re-cover and remained lower than age-matched nonlactatingcontrols, unlike areal BMD which was restored but re-mained slightly but not significantly lower than agematched controls (Tables 1 & 2).

Lactation is associated with uncoupled boneremodeling

Histomorphometry of the proximal tibia was done atd7Lac, d21Lac, and d21Post using calcein and demeclo-cycline labeling for trabecular bone surface. Bone forma-tion rate (BFR/BS) was increased by 33% between d7Lacand d21Post (P � .05) whereas eroded surfaces/bone sur-faces dropped dramatically from 2% at d7Lac to 0.37%at d21Post (P � .05) consistent with trends toward re-duced osteoclast number per bone perimeter (2.27 to 0.47;P � .069) and reduced percent osteoclast surface (5.85 to1.36; P � .09) (Figure 1B-C, Table SII).

Bone turnover markers generally confirmed thechanges seen by histomorphometry. Notably, serum C-telopeptide (Rat Laps) showed a marked increase at d7Lac(peak resorption) vs. 12 weeks old nonlactating controls(Figure 1D; P � .001) then returned to prelactation levelsat d21Lac and d21Post lactation. On the other hand,RANKL concentrations were unchanged between base-line and d7Lac of lactation (data not shown). Total serumosteocalcin showed a marked increase at d7Lac andd21Lac compared to 12wk old nonlactating controls (Fig-ure 1E; P � .001). For P1NP, another marker of boneformation, there was a statistically significant doubling atd7Lac (P � .001) compared to age-matched nonlactatingcontrol similar to the increase in C-telopeptide at d7Lac(Figure 1F). Serotonin, a hormone that regulates PTHrp inmammary tissue (9, 10), increased at d7Lac compared toage-matched controls (Table 3; P � .001). Serum scleros-tin at d21Lac was 50% lower in the lactating animalscompared to controls (Figure 1G; P � .001).

Gene expression was performed by whole bone plusmarrow analysis using qRT-PCR at, d7Lac and d21Lac aswell as d21Post compared to 12wk old nonlactating con-trol. Osterix and Atf4, markers of osteoblast differentia-tion, were markedly suppressed at d7Lac compared tobaseline (Figure 1I; P � .008, and P � .04). The osteoclas-togenesis inhibitory factor Opg was also down regulatedby 45% at d7Lac vs. baseline (Figure 1H; P � .002). Os-terix and Atf4 expression levels returned to normal by

Table 1. Body composition measured by DXA duringlactation in C57BL/6J mice.

d7Lac d21Lac d21Post

TotalaBMD(mg/cm2)Control 48.9 � 0.3 51.3 � 0.5 51.0 � 0.4

Lactating43.9 � 0.6* 45.0 � 1.1* 49.8 � 0.06bc

FemoralaBMD(mg/cm2)Control 59.7 � 0.9 61.6 � 0.8 63.9 � 0.8Lactating55.8 � 1.4 48.5 � 1.4*a 57.3 � 0.8*c

Bodyweight(g)Control 21.6 � 0.3 22.6 � 0.5 23.1 � 0.7Lactating27.1 � 0.4* 27.9 � 0.5* 24.8 � 0.4bc

% FatControl 14.8 � 0.3 14.9 � 0.8 18.3 � 1.8Lactating11.9 � 0.6 14.2 � 1.7 12.8 � 0.9*

Fat mass(g)Control 3.15 � 0.08 3.36 � 0.23 4.21 � 0.52Lactating3.14 � 0.18 3.90 � 0.43 3.08 � 0.24 (*P �

0.06)Lean

mass(g)Control 17.7 � 0.20 19.0 � 0.36 18.4 � 0.35Lactating23.2 � 0.47* 23.8 � 0.83* 20.9 � 0.41*bc

Values are mean � SEM n � 9–11 per group. Statistical analyses wereperformed using two-way ANOVA, P � 0.05; * lactating vs. age-matched controls (12wk vs. d7Lac, 15wk vs. d21Lac, 20wk vs.d21Post); a d7Lac vs. d21Lac, b *d7Lac vs. d21Post, c d21Lac vs.d21Post,

d12wk vs. 15wk, e 12wk vs. 20wk, f 15wk vs. 20wk.

4 Endocrinology


Tt.Ar,

d21Lac (Table SIII). The expression of osteoclast recruit-ment factor Rankl was unchanged for all time points (Ta-ble SIII). In respect to markers of fat metabolism in thebone marrow, Ppar�, a mediator of fatty acid oxidationand a regulator of FGF-21, and PDK4, a marker of markerof fatty acid metabolism were both slightly but not sig-nificantly increased at d7Lac (Table SIII). Despite themarked decrement in trabecular bone volume fraction, theabsolute volume of marrow adipose tissue using osmiumstaining with �CT of the tibia at four distinct sites dem-

onstrated a significant reduction at d21Lac vs. agematched controls for three sites; (VOI, volume of interest;adipocyte volume/total volume) 15wk control vs. d21Lac:VOI1 0.0047 � 0.0007 vs. 0.001 � 0.0002; VOI20.4874 � 0.05 vs. 0.0489 � 0.02; VOI4 0.0012 � 0.0001vs. 0.0002 � 0.0001 (Figure 2; P � .05).

Table 2. Femoral �CT phenotypes for lactating C57BL/6J mothers at day 7 (d7Lac), d21 (d21Lac) and d21 post-lactation (d21Post) compared to age-matched non-lactating controls (12wks, 15wks, 20wks).

12wk d7Lac 15wk d21Lac 20wk d21Post

DistalFemur

n � 5 n � 10 n � 5 n � 10 n � 5 n � 8

BV/TV(%)

6.3�0.43

4.4�0.42*

5.8 �0.32

3.3 �0.32*

6.0 � 0.38 2.7 �0.35*b

TbTh (�m) 40.3�1.3

35.8�1.0

38.9 �1.4

36.7� 1.3

41.9 �0.87

36.9 �1.5

TbN(1/mm)

3.7�0.05

3.3�0.12*

3.5 �0.05

2.9 �0.09*

3.3 � 0.10 2.6 �0.10*bc

TbSp (�m) 264.9�3.5

308.6�11.5

275.3� 3.7

336.4�10.1*

302.2 �9.9

393 �17.4*bc

SMI 2.1�0.08

3.3�0.09*

2.1 �0.10

3.6 �0.08*

2.1 � 0.05 3.5 �0.07*

Conn.D 62.1�5.9

33.4�6.8*

65.6 �3.6

20.6�5.5*

60.7 � 5.8 16.8 �3.1*

FemurMidshaft

Ct.Th (�m) 216.6�2.2

197.8�2.2*

218.6� 1.6

188.4�3.1*a

227.8 �3.0

212.1 �4.0bc

Ct.Ar/Tt.Ar(%)

49.5�0.42

45.1�0.62*

49.8 �0.35

42.6�1.0*

51.6 �0.58

47.4 �1.0*c

Tt.Ar(mm2)

1.73�0.01

1.85�0.03

1.75 �0.02

1.81�0.03

1.72 �0.01

1.83 �0.04

Ct.Ar(mm2)

0.86�0.01

0.83�0.01

0.87 �0.01

0.77�0.01*a

0.89 �0.01

0.86 �0.01c

Ma.Ar(mm2)

0.87�0.01

1.02�0.02*

0.88 �0.01

1.04�0.03*

0.83 �0.01

0.96 �0.04

Ct.Po (%) 7.9�0.13

9.0�0.06*

7.9 �0.05

9.2 �0.16*

7.9 � 0.19 8.6 �0.18c

Values are expressed as mean � SEM Statistical analyses were performed using two-way ANOVA, P � 0.05; * vs. age-matched non-lactatingcontrol (12wk vs. d7Lac, 15wk vs. d21Lac, 20wk vs. d21Post), a d7Lac vs. d21Lac, b d7Lac vs. d21Post, c d21Lac vs. d21Post, d 12wk vs. 15wk, e

12wk vs. 20wk, f 15wk vs. 20wk. �CT � micro-computed tomography; % BV/TV, Bone Volume/Total Volume; Tb.Th, Trabecular Thickness; Tb.N,Trabecular Number; Tb.Sp, Trabecular Separation; SMI, Structural Model Index; Conn.D, Connectivity Density; T.Ar, Total Area; Ct.Ar, CorticalArea; Ma.Ar, Marrow Area; Ct.Th, Cortical Thickness; Ct.Po, Cortical Porosity.



Body composition andmetabolic changes were presentduring and after lactation

Total body weight in grams washigher in the lactating dams vs. age-matched controls at d7Lac andd21Lac (Table 1). There was a 21%reduction in percent body fat atd7Lac in lactating mice vs. 12-weeknonlactating controls, however, per-cent body fat was not different atd21Lac. By d21Post lactation, per-cent body fat was 30% lower in thedams that had lactated vs. controls(Table 1; P � .05). Changes in per-cent fat were not due to changes inabsolute fat mass, which was not dif-ferent at d7Lac and d21Lac (Table1). Lean mass was similar fromd7Lac to d21Lac but higher thancontrols at all time points and thiswas associated with a marked in-crease in energy expenditure (EE;P � .005) in the lactating mice vs.controls (Figure S1A). At d21Postlactation, lean mass decreased signif-icantly compared to d7Lac andd21Lac mice, yet was still higherthan age-matched nonlactating con-trols. Adjusting EE for group andlean mass revealed no difference be-tween lactating and control micewhen a multiple regression modelwas used (r2 for lean � 0.65; P � .03;f value of model � 0.0018, FigureS1B). However, food consumptionof the lactating mice was markedlyincreased and total activity wasslightly reduced without a longitudi-nal change in total body weight, con-sistent with a shunting of nutrientstowards mammary tissue (FigureS1C-D).

To determine the mechanism forlactation-induced changes in bodycomposition we next investigatedthe effects of lactation on gene ex-pression in inguinal white adiposetissue (WAT), interscapular brownadipose tissue (BAT), and the mam-mary gland. In inguinal fat there wasglobal suppression of Pgc1� and

d7Lac d21Lac d21Post0

1

2

3

ES/B

S (%

)

*

Tota

l Ost

eoca

lcin

(ng/

mL)

Ctrl d7Lac d21Lac d21Post0

5

10

15

20

**

*


10

20

30

40

Rat

Laps

(ng/

mL)

*

a b

d7Lac d21Lac d21Post0

50

100

150

200

250

BFR

/BS

(µm3 /µ

m2 /y

ear)

*

*

CB

ED

GF


20

40

60

80

P1N

P (n

g/m

l)

a b

*a

*c*


500

1000

1500

Scle

rost

in (p

g/m

l)

*a

*

Rankl Opg0.0

0.5

1.0

1.5

Rel

ativ

e ex

pres

sion

at d

7 La

c

CtrlLac

*

Osterix Runx2 Atf40.0

0.5

1.0

1.5

Rel

ativ

e ex

pres

sion

at d

7 La

c

CtrlLac

* *

A

100 um 100 um 100 um 100 um 100 um 100 um

100 um 100 um 100 um 100 um 100 um 100 um

d7 d21 d21Post

Ctrl

Lac

H I

Figure 1. Bone turnover during lactation in controls and dams. Serum and boneswere harvested from lactating C57BL/6J females (n � 5–10) at d7Lac, d21Lac and d21Postlactation (black bars) were compared to 12wk old nonlactating controls (white bars). A,Representative �CT images of distal femur microarchitecture in lactating dams compared to age-matched nonlactating controls, scale bars � 100 �m, (B) Erosion Surface/Bone Surface (ES/BS)

6 Endocrinology


Ucp1, markers of brown-like adipogenesis, at d7Lac vs.12-week nonlactating control mice (Figure 3A-B; P �.002). This reduction persisted for Pgc1� at d21Lac butreturned to comparable expression levels relative to con-trol by d21Post lactation. Ucp1 mRNA expression levelsincreased 2 fold by d21Post lactation (Figure 3B). Pdk4,an enzyme that can promote fatty acid generation wasmore than 2 fold increased in inguinal fat relative to con-trol at d7Lac (P � .0001). In BAT, a similar pattern wasnoted; there was global suppression of Ucp1 mRNA (85%reduction, P � .0001) and Pgc1�, a transcriptional co-activator that modulates mitochondrial biogenesis and en-ergy homeostasis, was suppressed by 40% (P � .0002) atd7Lac vs. 12-week nonlactating control mice (FigureD-E). In mammary tissue, Ucp1 and Pgc1� mRNA wereup regulated significantly (P � .0001) at d7Lac whereasPgc1� transcripts were suppressed (Figure 3F-I; P � .002).

Due to the magnitude and acceleration of changes inbody composition associated with the high rates of boneresorption during lactation, we next investigated factorsthat might contribute to those effects. Undercarboxylatedosteocalcin, (Glu13 OCn), a peptide fragment that en-hances insulin sensitivity and secretion, was 25.1 � 1.2ng/ml at d7Lac, the peak of bone resorption. However, atd21Lac and then d21Post, Glu13 Ocn levels declined tolevels of 12.1 � 3.3 ng/ml (P � .01, d21Lac vs. d7Lac) and12.0 � 2.5 ng/ml respectively (P � .01, d21Post vs.

d7Lac). Serum leptin was not different at d7Lac, a timewhen fat mass was lost (Table 3). Adiponectin levels atd7Lac were nearly 50% lower than baseline levels (P �.001) reflecting the acute changes in adipose tissue; in con-trast serotonin concentrations increased significantly atd7Lac vs. controls (Table 3; P � .001).

Insulin concentrations were lowest at baseline andd7Lac but by d21Lac were increased significantly frombaseline: 0.445 � 0.066 vs. 0.729 � 0.119 (Table 3; P �.05). In the post lactation phase, serum insulin levels re-turned to baseline values. Insulin-like growth factor-I(IGF-I) serum levels increased nearly 50% from d7Lac tod21Post (P � .0002) and were greater than control atd21Post (Table 3; P � .01). Fibroblast growth factor-21(FGF-21), a counter-regulatory factor secreted by the liverduring states of substrate insufficiency, revealed a differ-ent pattern: serum levels of FGF-21 rose more than 2.4fold from baseline to d7Lac (P � .0001) and then re-mained elevated at d21Lac (P � .005) and d21Post (P �.0001), although less than d7Lac (Table 3). Consistentwith circulating changes, hepatic Fgf21 mRNA was in-creased 8 fold in the liver at d7Lac over baseline (P � .03,data not shown).

Due to changes in serum insulin during lactation andthe rise in endogenous insulin sensitizing peptides such asundercarboxylated osteocalcin and FGF-21, we per-formed an insulin tolerance test (ITT) at d7Lac in nonlac-

tating age-matched B6 controls (n �8) and lactating B6 mice (n � 6).Random fed glucose at d7Lac wassignificantly lower than control se-rum from age matched mice (Figure4A; P � .05). During ITT, blood glu-cose of lactating mice reached the na-dir sooner and the nadir was lower

Legend to Figure 1 Continued. . .histomorphometry in femur, (C) Bone Formation Rate/Bone Surface (BFR/BS) histomorphometryin femur, (D) RatLaps (C-telopeptide), (E) total osteocalcin, (F) P1NP, (G) sclerostin, (H) RelativemRNA expression of Rankl and Opg in tibia, (I) Relative mRNA expression of Osterix, Runx2 andAtf4 in tibia. Bars represent the mean � SEM. Statistical analyses were performed with Student’st test for B-C and H-I and ANOVA for D-G, P � .05, *compared to controls (* vs. d7Lac for ES/BSand BFR/BS); a represents d7Lac vs. d21Lac, b represents d7Lac vs. d21Post, c represents d21Lacvs. d21Post.

Table 3. Metabolic markers measured in lactating C57BL/6J mothers at day 7 (d7Lac), d21 (d21Lac) and d21 post-lactation (d21Post) compared to non-lactating controls (12wks).

12wk d7Lac d21Lac d21Post

Leptin (pg/mL) 1023.3 �202.6

989.7 �74.9

1310.9 �176.4

1147.7 �150.2

Adiponectin (ug/mL) 38.1 �1.4

22.5 �1.2*

32.6 �2.0a

25.9 �1.3*c

Serotonin (ng/mL) 2223.0 �67.6

3165.3 �185.9*

3197.7 �281.7*

3936.9 �356.6*

Insulin (ng/mL) 0.445 �0.06

0.563 �0.09

0.729 �0.11*

0.445 �0.06

IGF-1 (ng/mL) 411.3 �25.9

360.9 �41.9

445.9 �46.2

646.2 �39.6*bc

FGF-21 (pg/mL) 224.2 �8.17

551.0 �65.9*

447.6 �66.3*

505.4 �70.3*

Values are expressed as mean � SEM n � 5–11. Statistical analyses were performed using one-way ANOVA, P � 0.05; * vs. 12wk control; a d7Lacvs. d21Lac, b d7Lac vs. d21Post, c d21Lac vs. d21Post.



compared to nonlactating B6 mice (Figure 4B; P � .001).After day 14 of lactation, the increase in insulin sensitivity

was no longer present with a glucose pattern nearly iden-tical to age-matched controls (data not shown).

FGF-21 null mice (F21-KO) do not lose bone withlactation

To determine whether FGF-21 was a major mediator oflactation-induced changes in bone remodeling and energymetabolism, we next compared FGF-21 null (F21-KO)lactating dams at day 14 (d14Lac) to age-matched non-lactating virgin controls (Ctrl; 16wks old). Day 14 of lac-tation was chosen as a time point because a priori weexpected to see peak lactation-induced bone loss. Pup sizeand growth were similar between the two genotypesacross lactation and post weaning. �CT analyses revealedthat significant lactation-induced trabecular and corticalbone loss occurred in WT mice, but was not present inF21-KO mice (Figure 5A-D). Histomorphometry showed

1.0 mm 1.0 mm 1.0 mm 1.0 mm

Ctrl d21Lac

Figure 2. Lactation induces decline in marrow adiposity.Representative images of osmium staining for marrow adiposity.

C57BL/6J females at d21Lac were compared to 15wk old nonlactatingage-matched controls. Tibias were stained for marrow fat usingosmium tetraoxide (OsO4) and then imaged using uCT using a voxelsize of 10 �m. Scale bars � 1.0 mm.

A

D

G

CB

E

H

F

I

Ctrl d7Lac d21Lac d21Post0.0

0.5

1.0

1.5

WA

T P

gc1

/ H

prt /

B2M

*

bc

**


1

2

3

WA

T P

dk4

/ Hpr

t / B

2M* *

*


0.5

1.0

1.5

BAT

Ucp

1 / 1

8s

a

*

b

*


0.5

1.0

1.5

BAT

Pgc

1 /

18s

*

*

bc

*


0.5

1.0

1.5

2.0

MA

M U

cp1

/ Rpl

13a

*

*

b

*

*a*


0.5

1.0

1.5

2.0

MA

M P

gc1

/ Rp

l13a

**b

**

*

*a

*


1

2

3

WA

T U

cp1

/ Hpr

t / B

2M

* *

bc


1

2

3

MA

M P

dk4

/ Rpl

13a

*b

*

*b*

*

a


0.5

1.0

1.5

MA

M P

gc1

/ Rp

l13a

c

**

Figure 3. Lactation induces alterations in gene expression of metabolic markers in fat depots. Adipose tissue harvested fromlactating C57BL/6J females (n � 8–10) at d7Lac, d21Lac and d21Post lactation (black bars) were compared to 12wk old nonlactating controls(white bars). A–C, Inguinal white adipose tissue (WAT) mRNA expression of Pgc1�, Ucp1 and Pdk4, (D-E) Interscapular brown adipose tissue (BAT)mRNA expression of Ucp1 and Pgc1�, (F-I) Mammary (MAM) mRNA expression of Ucp1, Pgc1�, Pdk4 and Pgc1�. Bars represent the mean � SEM.Statistical analyses were performed with ANOVA, P � .05, *compared to controls; a represents d7Lac vs. d21Lac, b represents d7Lac vs. d21Post, c

represents d21Lac vs. d21Post.

8 Endocrinology


that osteoclast number and surface were significantly in-creased in d14Lac WT mice compared to age matchedcontrols, but not in F21-KO mice (Figure 5E; P � .005);similarly, osteoblast number and surface were decreasedon d14Lac compared to age-matched controls in WTmice, but not in F21-KO mice (Figure 5F; P � .001). Serum

CTX-1 (RatLaps) was increased on d14Lac in WT micecompared to age matched controls (P � .05), but this re-sponse was largely absent in F21-KO mice (Figure 5G).Serum P1NP on d14Lac was comparable to age matchedcontrols in WT mice despite the increased bone resorption;in contrast, P1NP was increased on d14Lac compared to

age matched controls in F21-KOmice (Figure 5H; P � .001).

Discussion

In the current study we demon-strated that lactation affects not onlyskeletal remodeling but also wholebody metabolism in mice. In addi-tion we noted several features of lac-tation in B6 mice that are potentiallyrelevant and translatable. First, boneloss in lactating mice is significantand involves both cortical and tra-becular elements, although each

Bloo

d gl

ucos

e (m

g/dL

)

Ctrl d7Lac0

50

100

150

200

*

Time (min)

Bloo

d gl

ucos

e (%

bas

elin

e)

0 20 40 60 80 100 1200

50

100

150Ctrld7Lac

* *

BA

Figure 4. Lactation induces increase in insulin sensitivity. Serum from lactatingC57BL/6J females (n � 5–10) at d7Lac (black bars) were compared to 12wk old nonlactating(white bars) for (A) Random-fed glucose, (B) Insulin tolerance test (ITT) in which blood glucosewas monitored for 120 minutes in random-fed control (dashed line) and d7Lac (solid line) micethat were injected IP with 1 U/kg insulin at time 0. Blood glucose values are expressed as %baseline. All data points represent the mean � SEM. Statistical analyses were performed withStudent’s t test, P � .05, *compared to controls.

Trab

ecul

ar B

V/T

V

WT F21-KO0.00

0.05

0.10

0.15

*

d14LacCtrl + +

Tb.N

(1/m

m)

WT0

1

2

3

4

5

*

++

WT F21-KO

0.0

0.2

0.4

0.6

*

Ob

.N/B

.Ar (

%)

WT F21-KO0

2

4

6

8

10

*

+ +

WT

F21-

KO

Ob

.S/B

.S (%

)

WT F21- KO0

5

10

15

*

+ +

Oc.

S/B

.S (%

)

WT F21-KO0

5

10

15*

+ +

Oc.

N/B

.Ar (

%)

WT F21- KO0

5

10

15

*

+ +

Seru

m C

TX-1

(ng/

mL)

WT F21- KO0

20

40

60

80

100 *

++*

Ctrl d14Lac Ctrld14Lac

F21-KO

Ctrld14Lac

Cor

tical

BV

/TV

Ctrld14Lac

Ctrld14Lac

Ctrld14Lac

WT F21- KO0

5

10

15

20*+

+

Seru

m P

1NP

(ng/

mL)

Ctrld14Lac

BA C D

FE

HG

Figure 5. FGF-21 is essential for lactation-induced skeletal and metabolic changes. FGF21-KO (F21-KO) mice and WT controls(16-week-old females, n � 5) were dissected on lactation day 14 (d14Lac). Bone and metabolic parameters on d14Lac (black bars) were comparedwith age-matched nonlactating virgin controls (Ctrl, white bars). (A-D), �CT analyses of proximal tibiae, scale bars � 1.0 mm, (E-F),Histomorphometry analyses of distal femurs, (G) Serum CTX-1, (H) Serum P1NP. Bars represent the mean � SEM. Statistical analyses wereperformed with Student’s t test, P � .05; * in B-H compares d14Lac with virgin controls in the same genotype; “�” in B-H compares F21-KOwith WT controls under the same physiological condition.



compartment responds distinctly. Remarkably, the globalloss of bone mass is associated with a compensatory in-crease in femoral cross sectional area, a mechanism thatlikely maintains bone strength during lactation and recov-ery (11–12, 25–27). A recent longitudinal study in humansconfirmed that repetitive and prolonged lactation protectsagainst hip fractures late in life, possibly due to this com-pensation in the cortical compartment (28). In contrast tothe cortex where bone mass recovered completely, trabec-ular bone was lost and never fully restored in the B6 mice,possibly because there is normally some concomitant age-related bone loss in B6 females during the observationperiod (29). It would be important to know, through theuse of high-resolution uCT imaging whether lactatingmothers have full trabecular bone recovery following lac-tation and whether the compensatory increase in femoralcross sectional area is responsible for the reduction in hipfractures late in life among women with repeated and pro-longed lactation periods.

The bone loss of lactation is due to both acceleratedbone resorption and suppressed bone formation. Themechanisms that lead to a reduction in bone formationduring the early phases of lactation are multifactorial.However, high levels of FGF-21, suppressed IGF-I, in-creased circulating serotonin or changes in local or circu-lating sclerostin might all contribute to a down regulationin bone formation. For example, serotonin regulates localPTHrP production in mammary tissue in response to el-evated prolactin levels (9, 10). Increased circulating sero-tonin from mammary tissue rather than from the gut couldlead to lower bone formation via inhibition of the Wntsignaling pathway. Sclerostin is released from osteocyteswithin the remodeling unit and could suppress bone for-mation at a time when calcium requirements in other tis-sues are very high (30). Increased circulating FGF-21 isanother possible cause of reduced bone formation sincetransgenic FGF-21 mice have markedly suppressed ratesof bone formation primarily due to the increased activityof PPAR� as well as suppression of Wnt signaling (31).

A second notable finding in our study was the skeletalrecovery of B6 mice post weaning and the relationship tochanges in metabolic factors. Interestingly, the restorationof bone mass is time and site specific such that total aBMDis recovered more rapidly than femoral aBMD, which maybe due to the contribution from trabecular elements in theproximal and distal femur to femoral aBMD measure-ments. Although skeletal recovery has been reported pre-viously after lactation, the association with systemic andlocal metabolic factors has not been described (32, 33).There were only modest changes in gene expression withinthe bone marrow for genes related to adiposity, but thesewere consistent with the decline in bone marrow adiposity

quantitated by osmium �CT in the lactating dams. Im-portantly, serum sclerostin levels declined at d21Lac, andthese changes were reflected in the histomorphometry atd21Lac and changes in biochemical markers of bone turn-over. In contrast to these data, another recent examinationof lactation found that sclerostin gene expression was re-duced somewhat later in the process of lactation, at d7post weaning (27).

Third, in this study we found that high levels of FGF-21are related to changes in body composition and skeletalremodeling. The high metabolic demand during lactationis also reflected by the reduction in marrow adipose tissueat d21Lac. At the same time there was a marked increasein circulating FGF-21 and an increase in insulin sensitivityat d7Lac. The timing of this sensitivity is remarkable inthat it also coincided with the greatest bone resorption (ie,d7Lac) during lactation. Both the Karsenty and Clemensgroups have shown that insulin secretion and sensitivityare increased during states of accelerated bone turnover inmice through release of undercarboxylated osteocalcinfrom the bone matrix (35–37). Undercarboxylated osteo-calcin (Glu13 Ocn) peaked at d7Lac vs. baseline and re-turned to control values at d21Lac and d21Post. Thus,insulin sensitivity during lactation could be related to highconcentrations of circulating FGF-21 and the release ofmatrix peptides such as undercarboxylated osteocalcin aswell as other hormonal factors.

In the FGF-21 null mice the cellular mechanism forskeletal resistance during the high calcium demand of lac-tation is likely related to a relative reduction in bone re-sorption compared to wild-type lactating mice. This wasconfirmed by the near absence of change in CTx levelsduring lactation in the FGF-21 null mice. Similarly it wasreported that in both male and female FGF-21 null mice,bone resorption was consistently reduced, and theRANKL/OPG ratio low, although the mechanism forthese changes was not well delineated (31). Moreover, it isuncertain why FGF-21 null osteoclasts do not respond tochanges in calciotropic hormones and the absence of es-trogen during the high calcium demands of lactation. In-terestingly, despite the absence of significant bone resorp-tion in the FGF-21 nulls, bone formation was notsuppressed (Figure 5F) suggesting an uncoupling in theremodeling unit favoring skeletal protection rather thanloss. It is conceivable that the FGF-21 null mice have adifferent pattern of sclerostin expression compared to lac-tating B6 mice, which have high levels during early lacta-tion but decrease by d21Lac. Alternatively, osteoclastnumber is not altered in FGF-21 null mice. Hence thesecells may continue to release coupling factors that enhancebone formation even though the osteoclasts are not ac-tively resorbing bone. Furthermore, we could not deter-

10 Endocrinology


mine from our studies whether osteocytic osteolysis dur-ing lactation was the chief mechanism for calcium releasefrom the skeleton in these mice, although this is a strongpossibility (26). Finally, changes in FGF-21 could also beimportant in the recovery phase post lactation. It is con-ceivable that a drop in FGF-21 could lead to a reboundenhancement in bone formation and subsequent decline inbone resorption, although several factors are likely in-volved. For example, circulating IGF-I was increased postlactation in C57BL/6J mice (33). Since FGF-21 can regu-late hepatic IGF-I through suppression of growth hor-mone (GH) mediated STAT 5 signaling, a decline inFGF-21 and an increase in IGF-I may combine to allowrecovery of BMD post lactation.

In conclusion, we established that whole body meta-bolic changes accompany uncoupling in bone remodelingduring lactation. Increased circulating FGF-21 affectswhole body metabolism, insulin sensitivity and skeletalremodeling during this period. Further studies are neededto define the mechanisms implicit in the FGF-21 skeletalresponse and to determine other skeletal factors that mayregulate whole body metabolism.

Acknowledgments

This work was supported by NIDDK R24 092759 to CJR, NIHK23 AR051483–06 to SAB, Canadian Institutes of Health Re-search #84253 to CK, NIAMS 5R01AR45548 to GK, NIDDKRO1DK089113, March of Dimes #6-FY13–137 and WelchFoundation I-1751 to YW. Additional support was provided bythe Maine Medical Center Research Institute BioinformaticsCore Facility as supported by NIH P20 GM103465 to DonM.Wojchowski and the Small Animal Imaging Core Facility assupported by NIH P30 GM103392 to Robert Friesel. We thankTerry Henderson, Ellen McMonagle, Elizabeth McMenaminand Kathleen Bishop for technical assistance and Katherine Mo-tyl for critical reading of the manuscript.

*Address all correspondence and requests for reprints to: Dr.Clifford Rosen, ME Medical Center Research Institute, 81 Re-search Drive, Scarborough, ME 04074–7205, Phone: 207–396-8228, Fax: 207–396-8174, Email: [email protected] (reprints);Dr. Yihong Wan, University of Texas Southwestern MedicalCenter, 6001 Forest Park Road, Room ND8.502B, Dallas, TX75390–9041, Phone: 214–645-6062, Fax: 214–645-6067,Email: [email protected].

8Current address Institut de recherches cliniques de Mon-tréal, Montréal, QC, Canada, H2W 1R7

9Current address Département de Médecine, Université deMontréal, QC, Canada, H3C 3J7

No disclosures.This work was supported by NIDDK R24 092759 to CJR,

NIH K23 AR051483–06 to SAB, Canadian Institutes of Health

Research #84253 to CK, NIAMS 5R01AR45548 to GK, NI-DDK RO1DK089113, Mar of Dimes #6-FY13–137 and WelchFoundation I-1751 to YW. Additional support was provided bythe Maine Medical Center Research Institute BioinformaticsCore Facility as supported by NIH P20 GM103465 to DonM.Wojchowski and the Small Animal Imaging Core Facility assupported by NIH P30 GM103392 to Robert Friesel.

References

1. Glasier A and McNally AS. Physiology of lactation. Bailliere’s Clin-ical Endocrinology and Metabolism. 1990;4:379–395.

2. Neville MC. The physiologic basis of milk secretion. Ann NY AcadSci. 1990;586:1–11.

3. Thomas E, Lee-Pullen T, Rigby P, Hartmann P, Xu J, Zeps N. Re-ceptor activator of NF-�B ligand promotes proliferation of a puta-tive mammary stem cell unique to the lactating epithelium. StemCells. 2012 Jun;30(6):1255–64.

4. Kovacs CS, Kronenberg HM. Maternal-fetal calcium and bone me-tabolism during pregnancy, puerperium, and lactation. Endocr Rev.1997;18(6):832–72.

5. Sowers M, Corton G, Shapiro B, Jannausch ML, Crutchfield M,Smith ML, Randolph JF, Hollis B. Changes in bone density withlactation. JAMA. 1993;269(24):3130–5.

6. Sowers M, Zhang D, Hollis BW, Shapiro B, Janney CA, CrutchfieldM, Schork MA, Stanczyk F, Randolph J. Role of calciotrophic hor-mones in calcium mobilization of lactation. Am J Clin Nutr. 1998;67(2):284–91.

7. Sowers M, Eyre D, Hollis BW, Randolph JF, Shapiro B, JannauschML, Crutchfield M. Biochemical markers of bone turnover in lac-tating and nonlactating postpartum women. J Clin EndocrinolMetab. 1995;80(7):2210–6.

8. Ardeshirpour L, Brian S, Dann P, VanHouten J, Wysolmerski J.Increased PTHrP and decreased estrogens alter bone turnover but donot reproduce the full effects of lactation on the skeleton. Endocri-nology. 2010 Dec;151(12):5591–601.

9. Hernandez LL, Gregerson KA, Horseman ND. Mammary glandserotonin regulates parathyroid hormone-related protein and otherbone-related signals. Am J Physiol Endocrinol Metab. 2012 Apr;302(8):E1009–1515.

10. Collieer RJ, Hernandez LL, Horseman ND. Serotonin as a homeo-static regulator of lactation. Domest Anim Endocrinol. 2012;43:161–170.

11. Kirby BJ, Ma Y, Martin HM, Buckle Favaro KL, Karaplis AC,Kovacs CS. Upregulation of calcitriol during pregancy and skeletalrecovery after lactation do not require PTH. J Bone Min Res. 2013;28:1987–2000.

12. Liu XS, Ardeshirpour L, VanHouten JN, Shane E, Wysolmerski JJ.Site-specific changes in bone microarchitecture, mineralization, andstiffness during lactation and after weaning in mice. J Bone MinerRes. 2012 Apr;27(4):865–75.

13. Kirby BJ, Ardeshirpour L, Woodrow JP, Wysolmerski JJ, Sims NA,Karaplis AC, Kovacs CS. Skeletal recovery after weaning does notrequire PTHrP. J Bone Miner Res. 2011 Jun;26(6):1242–51.

14. Carneiro RM, Prebehalla L, Tedesco MB, Sereika SM, Hugo M,Hollis BW, Gundberg CM, Stewart AF, Horwitz MJ. Lactation andbone turnover: a conundrum of marked bone loss in the setting ofcoupled bone turnover. J Clin Endocrinol Metab. 2010 Apr;95(4):1767–76. Epub 2010 Feb 11.

15. VanHouten JN, Dann P, Stewart AF, Watson CJ, Pollak M, KaraplisAC, Wysolmerski JJ. Mammary-specific deletion of parathyroidhormone-related protein preserves bone mass during lactation.J Clin Invest. 2003;112(9):1429–36.

16. Potthoff MJ, Inagaki T, Satapati S, Ding X, He T, Goetz R, Mo-



mailto:[email protected]

mailto:[email protected]

hammadi M, Finck BN, Mangelsdorf DJ, Kliewer SA, Burgess SC.FGF-21 induces PGC-1alpha and regulates carbohydrate and fattyacid metabolism during the adaptive starvation response. Proc NatlAcad Sci U S A. 2009;106:10853–10858.

17. DeMambro VE, Clemmons DR, Horton LG, Bouxsein ML, WoodTL, Beamer WG, Canalis E, Rosen CJ. Gender-specific changes inbone turnover and skeletal architecture in igfbp-2-null mice. Endo-crinology. 2008;149(5):2051–61.

18. Scheller EL, Troiano N, VanHoutan JN, Bouxsein MA, Fretz JA, XiY, Nelson T, Katz G, Berry R, Church CD, Doucette CR, RodehefferMS, MacDougald OA, Rosen CJ, Horowitz MC. Use of osmiumtetroxide to visualize and quantify marrow adipocytes. Methods inEnzymology. (2013) Methods in Adipose Tissue Biology. Febru-ary,2014;Ed by Macdougald, O. Elsevier Press,p 360.

19. Dempster DW, Compston JE, Drezner MK, Glorieux FH, Kanis JA,Malluche H, Meunier PJ, Ott SM, Recker RR, Parfitt AM. Stan-dardized nomenclature, symbols, and units for bone histomorphom-etry: a 2012 update of the report of the ASBMR HistomorphometryNomenclature Committee. J Bone Miner Res. 2013 Jan;28(1):2–17).

20. Ferron M, Wei J, Yohizawa T, Ducy P, Karsenty G. An ELISA basedmethod to quantify osteocalcin carboxylation in mice. Biochem Bio-phys Res Commun. 2010 Jul;397(4):691–696.

21. Motyl KM Bishop KA, DeMambro VE, Bornstein SA, Le P, KawaiM, Lotinun S, Horowitz MC, Baron R, Bouxsein ML, Rosen CJ.Altered thermogenesis and impaired bone remodeling in Misty mice.J Bone Miner Res. 2013 Sep;28(9):1885–97.

22. Kaiyala KJ, Schwartz MW. Toward a more complete (and less con-troversial) understanding of energy expenditure and its role in obe-sity pathogenesis. Diabetes. 2011 Jan;60(1):17–23.

23. Kaiyala KJ, Morton GJ, Leroux BG, Ogimoto K, Wisse B, SchwartzMW. Identification of body fat mass as a major determinant ofmetabolic rate in mice. Diabetes. 2010 Jul;59(7):1657–66.

24. Weitgasser R, Gappmayer B. Newer Portable Glucose Meters-An-alytical Improvement Compared with Previous Generation Devices.Clinical Chemistry. 1999;45:1821–1825.

25. Colleran HL, Wideman L, Lovelady CA. Effects of energy restrictionand exercise on bone mineral density during lactation. Med SciSports Exerc. 2012 Aug;44(8):1570–9.

26. Qing H, Ardeshirpour L, Pajevic PD, Dusevich V, Jähn K, Kato S,Wysolmerski J, Bonewald LF. Demonstration of osteocytic perila-cunar/canalicular remodeling in mice during lactation. J Bone MinerRes. 2012 May;27(5):1018–29.

27. Collins JN, Kirby BJ, Woodrow JP, Gagel RF, Rosen CJ, Sims NA,Kovacs CS. Lactating Ctcgrp nulls lose twice the normal bone min-eral content due to fewer osteoblasts and more osteoclasts, whereasbone mass is fully restored after weaning in association with up-regulation of Wnt signaling and other novel genes. Endocrinology.2013 Apr;154(4):1400–13.

28. Bjørnerem A, Ahmed LA, Jørgensen L, Størmer J, Joakimsen RM.Breastfeeding protects against hip fracture in postmenopausal wom-en: the Tromsø study. J Bone Miner Res. 2011 Dec;26(12):2843–50.

29. Turner CH, Hsieh YF, Müller R, Bouxsein ML, Baylink DJ, RosenCJ, Grynpas MD, Donahue LR, Beamer WG. Genetic regulation ofcortical and trabecular bone strength and microstructure in inbredstrains of mice. J Bone Miner Res. 2000 Jun;15(6):1126–31.

30. Keller H, Kneissel M. SOST is a target gene for PTH in bone. Bone.2005 Aug;37(2):148–58.

31. Wei W, Dutchak PA, Wang X, Ding X, Wang X, Bookout AL, GoetzR, Mohammadi M, Gerard RD, Dechow PC, Mangelsdorf DJ,Kliewer SA, Wan Y. Fibroblast growth factor 21 promotes bone lossby potentiating the effects of peroxisome proliferator-activated re-ceptor �. Proc Natl Acad Sci U S A. 2012 Feb; 109(8):3143–3148821.

32. Ardeshirpour L, Dann P, Adams DJ, Nelson T, VanHouten J,Horowitz MC, Wysolmerski JJ. Weaning triggers a decrease in re-ceptor activator of nuclear factor-kappaB ligand expression, wide-spread osteoclast apoptosis, and rapid recovery of bone mass afterlactation in mice. Endocrinology. 2007;148(8):3875–86.

33. Bowman BM, Siska CC, Miller SC. Greatly increased cancellousbone formation with rapid improvements in bone structure in the ratmaternal skeleton after lactation. J Bone Miner Res. 2002;17(11):1954–60.

34. Lundåsen T, Hunt MC, Nilsson LM, Sanyal S, Angelin B, AlexsonSE, Rudling M. PPARalpha is a key regulator of hepatic FGF-21.Biochem Biophys Res Commun. 2007 Aug 24;360(2):437–40.Epub 2007 Jun 2.

35. Ferron M, Wei J, Yoshizawa T, Del Fattore A, DePinho RA, Teti A,Ducy P, Karsenty G. Insulin signaling in osteoblasts integrates boneremodeling and energy metabolism. Cell. 2010 Jul;142(2):296–30823.

36. Fulzele K, Clemens TL. Novel functions for insulin in bone. Bone.2012 Feb;50(2):452–6.

37. Clemens TL, Karsenty G. The osteoblast: an insulin target cell con-trolling glucose homeostasis. J Bone Miner Res. 2011 Apr;26(4):677–80.

12 Endocrinology


FGF-21 and skeletal remodeling during and after lactation in C57BL/6J mice

Documents

Transcript of FGF-21 and skeletal remodeling during and after lactation in C57BL/6J mice