fragmented sleep accelerates tumor growth and progression ...€¦ · 1 q1 integrated systems and...

1 Integrated Systems and TechnologiesQ1

2 Fragmented Sleep Accelerates Tumor Growth and3 Progression through Recruitment of Tumor-Associated4 Macrophages and TLR4 SignalingQ256 Fahed Hakim1, Yang Wang1, Shelley X.L. Zhang1, Jiamao Zheng1, Esma S. Yolcu2, Alba Carreras1,7 Abdelnaby Khlayfa1, Haval Shirwan2, Isaac Almendros1, and David Gozal1

8 Abstract9 Sleep fragmentationQ5 (SF) is a highly prevalent condition and a hallmark of sleep apnea, a condition that has10 been associated with increased cancer incidence and mortality. In this study, we examined the hypothesis that11 sleep fragmentation promotes tumor growth and progression through proinflammatory TLR4 signaling. In the12 design, we compared mice that were exposed to sleep fragmentation one week before engraftment of syngeneic13 TC1 or LL3 tumor cells and tumor analysis three weeks later. We also compared host contributions through the14 use of mice genetically deficient in TLR4 or its effector molecules MYD88 or TRIF. We found that sleep15 fragmentation enhanced tumor size and weight compared with control mice. Increased invasiveness was16 apparent in sleep fragmentation tumors, which penetrated the tumor capsule into surrounding tissues including17 adjacent muscle. Tumor-associated macrophages (TAM) were more numerous in sleep fragmentation tumors18 where they were distributed in a relatively closer proximity to the tumor capsule, compared with control mice.19 Although tumors were generally smaller in both MYD88�/� and TRIF�/� hosts, the more aggressive features20 produced by sleep fragmentation persisted. In contrast, these more aggressive features produced by sleep21 fragmentationwere abolished completely inTLR4�/�mice. Ourfindings offermechanistic insights into how sleep22 perturbations can accelerate tumor growth and invasiveness through TAM recruitment and TLR4 signaling23 pathways. Cancer Res; 1–9. �2014 AACR.242526

27 Introduction28 In recent years, the possibility that sleep duration and29 overall sleep characteristics may affect overall cancer out-30 comes has been advanced (1). Indeed, in several epidemi-31 ologic studies spanning the last decade, the presence of32 altered sleep duration, both shortened and prolonged sleep,33 has been associated with higher incidence or adverse34 prognosis for several solid tumors. (2–17). However,35 although the role of the circadian clock system in tumor-36 igenesis has been extensively explored (18, 19), no animal37 models have thus far examined whether the association38 between disrupted sleep and tumorigenesis is indeed reca-39 pitulated, and if so, what potential mechanisms may under-40 lie such associations.41 In this context, some efforts to explore causal associations42 between a highly prevalent sleep disorder, namely obstruc-

44tive sleep apnea (OSA), and cancer have also taken place (20,4521), and have operated under the assumption that the46intermittent hypoxemia that characterizes patients with47OSA during their sleep period is likely to mimic the biologic48events that drive tumor growth (1, 22–27). The major find-49ings from these initial studies indicate that the periodic50oscillations in overall oxygenation during sleep in patients51with OSA impose overall adaptive changes in the tumor52metabolic cellular substrate that enhances their proliferative53and invasiveness properties (28). However, these studies54failed to explore another hallmark characteristic of OSA,55namely sleep fragmentation (SF), that is, the presence of56recurrent arousals aimed at restoring airflow that lead to57sleep discontinuity.58Using a similar logical paradigm, we hypothesized that59chronic sleep fragmentation, a very frequent occurrence in60many human disorders, including OSA, would be associated61with altered solid tumor proliferation and invasiveness in a62murine model (29, 30). Furthermore, we posited that sustained63sleep fragmentation would promote changes in the phenotypic64distribution of tumor-associated macrophages (TAM). Indeed,65TAMs have been identified as critically important constituents66of cancer microenvironment, and are major contributors to67cancer progression by releasing a vast array of growth factors,68cytokines, inflammatory mediators, and proteolytic enzymes69that underlie key components of tumor growth and invasion70(31, 32).

Authors' Affiliations: 1Pediatric Sleep Medicine, Department of Pediat-rics, Comer Children's Hospital, The University of Chicago, Chicago,Illinois; and 2Department of Microbiology and Immunology, The Universityof Louisville, Louisville, KentuckyQ3

Corresponding Author:David Gozal, Department of Pediatrics, Universityof Chicago, 5721 S. Maryland Avenue, MC 8000, Suite K-160, Chicago, IL60637. Phone: 773-702-6205; Fax: 773-702-4523; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-13-3014

�2014 American Association for Cancer Research.

AU

CancerResearch

www.aacrjournals.org 1

73 Materials and Methods74 Animals75 Male C57/B6, TLR4�/�, MYD88�/�, and TRIF�/� mice,76 weighing approximately 25 g, were purchased from The Jack-77 son Laboratory, housed in a 12 hours light/dark cycle (light on78 7:00 am to 7:00 pm) at a constant temperature (24�C � 1�C)79 and allowed access to food and water ad libitum. The exper-80 imental protocols were approved by the Institutional Animal81 Use and Care Committee and are in close agreement with the82 NIH Guide in the Care and Use of Animals. All efforts were83 made to minimize animal suffering and to reduce the number84 of animals used.

85 Sleep fragmentation86 The custom-designed sleep fragmentation approach used to87 induce sleep fragmentation in rodents has been previously88 reported in detail (29, 33), and relies on automated intermittent89 tactile stimulation of freely behaving mice in a standard90 laboratory mouse cage, using a near-silent motorized mechan-91 ical sweeper. Thismethod obviates the need for human contact92 and intervention, and does not involve introduction of foreign93 objects or touching of the animals during sleep. To induce sleep94 fragmentation, we chose a 2-minute interval between each95 sweep, implemented during the light period (7:00 a.m. to 7:0096 p.m.). Of note, 4 to 5 mice were housed in each cage to prevent97 the isolation stress.

98 Tumor cell lines and culture medium99 Weused TC-1 cells (ATCCCRL-2785) and 3LLC (ATCC CRL-100 1642) in all experiments. Both cell lines are derived from101 primary lung epithelial cells of C57/B6 mice and were cultured102 at 37�C, 95% air, 5% CO2 incubator in full tumor medium as103 recommended by the American Type Culture Collection104 (ATCC). TC-1 cells were cultured in RPMI-1640 medium with105 2 mmol/LQ6 L-glutamine adjusted to contain 1.5 g/L sodium106 bicarbonate, 10 mmol/L HEPES, and 1.0 mmol/L sodium107 pyruvate supplemented with 2 mmol/L nonessential amino108 acids, penicillin and streptomycin, and 10%FBS, all supplied by109 Gibco, Life Technologies, and selected with G418 (Gibco, Life110 Technologies) and hygromycin B (Calbiochem, EMDMillipore111 Corporation). 3LL cells were cultured in Dulbecco's Modified112 Eagle Medium (DMEM) supplemented with 10% FBS, L-gluta-113 mine, penicillin, and streptomycin, all supplied from (Gibco,114 Life Technologies). Study mice were inoculated with TC-1 or115 3LL cells (1 � 105 cells in 0.2 mL PBS) by subcutaneous116 injection into the right lower flank or right thigh for selected117 experiments.

118 Tumor model119 All mice strains, both in sleep conditions (SC) and sleep120 fragmentation, were housed at the same environmental con-121 ditions in the same cages. After 7 days of exposures, both sleep122 fragmentation and sleep conditionsmice were inoculated with123 TC-1 or 3LLC cells and tumor growth was monitored two to124 three times per week using a precision caliper. Average tumor125 sizewas calculated bymeasuring twoperpendicular diameters.126 Under these conditions, all the mice develop palpable tumors127 within 9 to 12 days. Animals bearing tumorswere euthanized at

129day 28 after injection; tumors were enucleated, scaled, and130subjected to different experiments.

131Matrix metalloproteinase in vivo live imaging132Twelve Q7C57/B6malemice (6 sleep fragmentation and 6 sleep133conditions) were injected subcutaneously with 1� 105 live TC1134cells into the lateral aspect of the right thigh using the same135protocol previously described; 28 days after tumor cell inoc-136ulation, mice were intravenously injected with 2 nmol of the137matrix metalloproteinase (MMP) probe MMPSense 750 FAST138(PerkinElmer; Product number: NEV10168), this MMP activa-139table agent is optically silent upon injection and produces140fluorescent signal after cleavage by disease-related MMPs.141Activation can occur by a broad range of MMPs, including142MMP 2, 3, 7, 9, 12, and 13. Two days before the MMPSense143injection, all the fur lining the tumor area was removed by hair144removal cream, after which mice were imaged 6, 12, and 24145hours after injection using the Xenogen IVIS Spectrum146(PerkinElmer) at the University of Chicago Optical Imaging147Core Facility. The tumor near IR fluorescence (average radiant148efficiency) of images was quantified using Living Image Soft-149ware (PerkinElmer).

150Local tumor invasiveness151To assess the differences in local tumor invasiveness152between the sleep fragmentation and sleep conditions groups,1537 days after sleep fragmentation/sleep conditions initiation,154C57/B6 mice were injected subcutaneously with TC1 cells into155the lateral aspect of the right thigh using 1� 105 cells in 0.2 mL156PBS. Twenty-eight days after tumor cell inoculation, both sleep157fragmentation and sleep conditions mice were euthanized and158subjected to wide resection of the tumor tissue with its159adjacent muscular and bone tissue of the thigh. The whole160specimen was fixed in 4% PFA Q8after it was cut into equivalent161size pieces. Those specimens were embedded in paraffin and162followed by cutting of 3 to 5 mmsections and were stained with163hematoxylin and eosin (H&E).

164Analysis of tumor-infiltrating cells165Tumors were mechanically disrupted in small pieces and166maintained overnight in TC1 complete growth medium but167without Geneticin (G418). After 12 hours, cells were harvested168and filtered through a 100 mmol/L nylon mesh cell strainer169(352350; BD Falcon). Before cell identification, viable cells were170selected by using the aqua-fluorescent reactive dye (L34957;171Invitrogen).TAMsweredefinedasCD45þ, CD11bþ, F4/80þ cells.

172Isolation of TAMs173Tumors were mechanically disrupted and incubated in174collagenase IV solution 1 mg/mL for 1 hour at 37�C. Cells175were filtered through a 100 Q9mm nylon mesh cell strainer and176CD11bþ cells were isolated by magnetic labeling following the177manufacturer's procedure (EasySep Mouse CD11b Positive178Selection Kit; StemCell, 18770).

179Immunofluorescence analysis180Tumors were excised, frozen in optimum cutting temper-181ature (OCT Q10), and stored in�80�C. Of Q11note, 10 mm cryosections

Hakim et al.

Cancer Res; 2014 Cancer Research2

184 were cut and stained with F4/80. Sections were washed several185 times in PBS, and blocked with a PBS/0.4% Triton X-100/0.5%186 TSA (tyramide signal amplification; PerkinElmer Life Sciences)187 blocking reagent/10% normal horse serum for 1 hour. SectionsQ12

188 were then serially incubated with a rabbit anti-mouse F4/80189 antibody (1:500; cat # 122604; BioLegendQ13 ) at 4�C for 24 hours190 and then washed in PBS for six times giving 5 minutes for191 each wash. Sections were incubated at room temperature for 1192 hour in biotinylated antibody (1:400; Vector Labs) in a PBS/193 0.4% TSA blocking reagent/10% horse serum solution, and194 then with streptavidin–horseradish peroxidase diluted at195 1:100 in PBS/0.5% TSA blocking reagent. Subsequently, the196 sections were incubated with TSA fluorescein reagents197 diluted at 1:50 in amplification diluent (PerkinElmer Life198 Sciences) for 2 minutes. Sections were then washed and199 mounted onto glass slides. Sections were visualized using a200 fluorescent microscope by an investigator who was blinded201 to the sample source.

202 Endocan staining203 Paraffin sections of sleep conditions and sleep fragmenta-204 tion tumor tissue were stained immunohistochemically for205 endocan using anti-mouse Endocan (cat # LIA-0905; Lungin-206 nov) as recommended by the manufacturer.

207 Total RNA isolation and gene expression208 Sorted TAMs were instantly frozen in liquid nitrogen after209 counting the amount of sorted cells. Total RNA was isolated210 using automated RNA extraction (Promega) and DNase-trea-211 ted according to the manufacturer's protocol. The RNA quan-212 tity and integrity were determined using a NanoDrop Spec-213 trophotometer and Agilent 2100 Bioanalyzer Nano 6000 Lab214 Chip assay (Agilent Technologies). Quantitative real-time PCR215 (qRT-PCR) was performed using the ABI 7500 instrument216 (Applied Biosystems). The cDNA was synthesized using a217 High-Capacity cDNA Archive Kit (Applied Biosystems). OfQ14

218 note, 500 ng of total RNA from both sleep fragmentation and219 sleep conditions samples were used to generate the cDNA220 templates for RT-PCR. TaqMan Master Mix Reagent Kit221 (Applied Biosystems) was used to amplify and quantify the222 transcripts in 20 mL reactions. Duplicate PCR reactions were223 performed in 96-well plates in parallel with the b-actin rRNA as224 a housekeeping gene. The steps involved in the reaction225 program included: the initial step of 2 minutes at 50�C;226 denaturation at 95�C for 10 minutes, followed by 45 thermal227 cycles of denaturation (15 seconds at 95�C) and elongation (1228 minutes at 60�C). Expression values were obtained from the229 cycle number (Ct value) using the Biosystems analysis software.230 TheseQ15 Ct values were averaged and the difference between231 the b-actin Ct (Avg) and the gene of interest Ct (Avg) was232 calculated (Ct-diff). The relative gene expression was analyzed233 using the 2�DDCt method. Quantitative results were expressed234 as the mean � SD. TheQ16 following TaqMan primer and235 probes were purchased from Applied Biosystems: interleukin236 (IL)-12b assay# Mm00434174_m1, inducible nitric oxide237 synthase (iNOS) assay# Mm00440502_m1, INF-g assay#238 Mm01168134_m1, Arginase1 assay# Mm00475988_m1, resis-239 tin-like a (Fizz1) assay# Mm00445109_m1, Mrc1 assay#

241Mm00485148_m1, IL-10 assay# Mm00439614_m1, TGF-b242assay# Mm01178820_m, TRL2 assay#Mm00442346_m1, TLR4243assay# Mm00445273_m1, and TLR6 assay# Mm02529782_s1.

244Statistical analysis245All data are reported as mean� SE. Comparisons for tumor246growth and all other time-course experiments among sleep247fragmentation and sleep conditionswere performed using one-248way ANOVA followed by unpaired Student t test with Bonfer-249roni correction. Comparison of all other quantitative data250between sleep fragmentation and sleep conditions was per-251formed using unpaired Student t tests. For all comparisons, a P252< 0.05 was considered as statistically significant.

253Results254Sleep fragmentation induces accelerated tumor size255growth256C57/B6 mice exposed to sleep fragmentation starting 1-257week before flank injection with TC-1 cells showed accelerated258tumor size growth with significant differences emerging at day25923 following injection and thereafter when compared with260mice under control sleep conditions (Fig. 1A; n ¼ 53/group;261sleep fragmentation vs. sleep conditions: P < 0.007). Tumor262weight was significantly higher at day 28 in sleep fragmenta-263tion–exposed mice (SF-57/B6: 1.955 � 0.680 g vs. SC-C57/B6:2641.046 � 0.479 g; P < 0.001). Similarly, mice (n ¼ 15/group)265exposed to sleep fragmentation and injected with 3LLC cells266showed significantly accelerated tumor growth (P < 0.001) and267tumor weight (P < 0.027) compared with sleep conditions (SF-26857/B6: 1.22 � 0.281 g vs. SC-C57/B6: 0.543 � 0.163 g; P <2690.027; Fig. 1B).

270Sleep fragmentation induces increased tumor271invasiveness272When TC-1 cells were injected subcutaneously in the273thigh area, increased invasiveness was apparent in SF-274C57/B6 tumors compared with SC-C57/B6 tumors, with275obvious penetration of surrounding tumor capsule and276extension/infiltration into adjacent muscle (Fig. 2A; n ¼27712/group; x2: P < 0.001). To further explore the increased278invasiveness of tumors in sleep fragmentation–exposed279mice, we performed live MMP imaging, which showed280significantly increased MMP activity in SF-C57/B6 tumors281vs. CS-C57/B6 (n ¼ 6; P < 0.01; Fig. 2B).

282TAM counts and distribution283When compared with CS-C57/B6 tumors, significant284increases in TAM counts occurred in SF-C57/B6 tumors285whether expressed as TAM/g tumor tissue (Fig. 3A; P <2860.01) or as TAM/whole tumor (Fig. 3B; P < 0.01). TAM in SF-287C57/B6 tumors were preferentially distributed in close prox-288imity to the tumor capsule, as compared with increased tumor289core location in tumors from SC-C57/B6 mice (Fig. 3D and E).290When endocan staining was used to identify new vessel291formation in the tumor capsule, SF-C57/B6 tumors had higher292expression of endocan when compared with sleep conditions293(Fig. 3F).

Disrupted Sleep and Cancer

www.aacrjournals.org Cancer Res; 2014 3

296 Sleep fragmentation induces TAM polarity shift toward297 M2 and higher TLR4 expression298 ToQ19 furtherunderstandwhether changes inmacrophagepolar-299 ity occurred during sleep fragmentation, we performed fluores-300 cence-activated cell sorting (FACS) of tumors for TAM (i.e.,301 CD45þ, CD11bþ, and F4/80þ) and further assessed the propor-302 tion of those expressing CD 86high, a M1marker or CD 206high, a303 M2 marker. A shift toward increased M2 marker expression in304 TAM of SF-C57/B6 tumors emerged, particularly in peripheral305 areasof the tumor(P<0.04; Fig. 4A).TAMshift inpolarity insleep306 fragmentation tumors was also apparently based on increased307 transcriptional expression of M2 markers when compared with308 sleep conditions, as evidenced by higher expression of Fizz1,309 Arg1, and Mrc1 (P < 0.01; Fig. 4B). TAM from sleep fragmenta-310 tion–exposed mice also expressed higher levels of TLR4 com-311 paredwith sleep conditions–exposedmice (P< 0.02; Fig. 4C), but312 nochanges inTLR2orTLR6wereapparent (datanot shown).On313 the basis of such findings, we examined TLR4 signaling as a314 potential pathway mediating sleep fragmentation–induced dif-315 ferences in tumorigenesis as described heretofore.

316 TLR4 signaling mediates sleep fragmentation–induced317 tumor progression318 When compared with CS-C57/B6 mice, significant reduc-319 tions in tumor size and weight occurred in SC-TLR4�/� mice

321(P < 0.001; Fig. 4D and E). More interestingly, the accelerated322growth and increases in size induced by sleep fragmentation323were completely abrogated in SF-TLR4�/� mice (Fig. 4D324and E; P < 0.05 vs. SC-TLR4�/� mice). These effects were325accompanied by significant reductions in TAM count in SF-326TLR4�/� mice (P < 0.01; Fig. 4F). These results prompted327further exploration as to whether TLR4 major downstream328signaling pathways, namely MYD88 or TRIF were particu-329larly and specifically recruited in sleep fragmentation. As330shown in Fig. 4G, tumors enucleated from SF-MYD88�/� and331SF-TRIF�/� mice showed reductions in tumor weight com-332pared with SF-C57/B6 (P < 0.03, P < 0.01, respectively), but333were still significantly larger than in SF-TLR4�/� mice. In334addition, small, albeit statistically significant differences335between sleep fragmentation and sleep conditions for336MYD88 and TRIF null mice remained.

337Discussion338This study shows that chronically fragmented sleep, a highly339prevalent condition associated with a multiplicity of human340disorders, leads to accelerated tumor growth and invasiveness341in mice and that such adverse effects are mediated, at least in342part, by changes in TAM polarity and TLR4 signaling. Indeed,343increases inM2macrophagemarkers were apparent in tumors

ATC1 Tumors

i i

ii

ii iii

iv

4,000

3,000

2,000

1,000

0

4,000

3,000

2,000

1,000

0

2,000

1,500

1,000

500

0

v

3LLC TumorsB

12 16 20

Days

Tu

mo

r vo

lum

e (

mm

3)

Tu

mo

r vo

lum

e (

mm

3)

3,000

2,000

1,000

0

Tu

mo

r w

eig

ht

(g)

1,500

1,000

500

0

Tu

mo

r w

eig

ht

(g)

2,000

1,500

1,000

500

0

Tu

mo

r vo

lum

e (

mm

3)

SC

SC1 SC2 SC3 SC4 SC5 SC6 SC7 SC8

SF

SCSF

SC SF

SF1 SF2 SF3 SF4 SF5 SF6 SF7 SF8

SC SF

SCSF

24 28 12 16 20

Days

24 28 12 16 20

Days

* *

**

24 28 12 16 20

Days

24 28

Figure 1. Sleep fragmentation (SF) induces accelerated tumor size growth both in TC1 and 3LLC tumors. A, C57/B6 mice exposed to sleep fragmentationstarting 1 week before flank injection with TC-1 cells showed accelerated tumor size growth and weight with significant differences when comparedwith mice under sleep conditions (SC). (i) The trajectory of tumor growth measured under sleep conditions (n ¼ 53), compared with tumors from sleepfragmentation–exposed mice; (ii) Sleep fragmentation, n ¼ 55; (iii) when comparing the average tumor growth for both conditions, statisticallysignificant differences emerged at day 23 following injection and sustained thereafter (P < 0.007). (iv) Macroscopic differences between the sleepfragmentation and sleep conditions tumors in samples of enucleated tumors. (v) Tumor weight was also significantly higher at day 28 in sleep fragmentation–exposedmice (SF-57/B6: 1.955� 0.680 g vs. SC-C57/B6: 1.046� 0.479 g; P < 0.001; n¼ 109 in each group). Of note, those experiments were performed inbatches of 20 mice each, using the same tumor cell injection and follow-up protocols. B, mice (n ¼ 15/group) exposed to sleep fragmentation and injectedwith 3LLC cells showed significantly accelerated tumor growth comparedwith sleep conditions, (i) with differences emerging after 19 days from cells injection(P < 0.001), and (ii) significant tumor weight differences at enucleation (SF-57/B6: 1.22 � 0.281 g vs. SC-C57/B6: 0.543 � 0.163 g; P < 0.027). �, P < 0.05.Q17

Hakim et al.


346 of sleep fragmentation–exposed mice, and were accompanied347 by increased in vivo and in vitro MMP activity (34, 35). We348 further show that similar to previous reports, TLR4 signaling349 not only underlies significant components of tumor progres-350 sion (34, 35), but also seems to mediate the differences in351 tumor proliferation between sleep fragmentation and sleep352 conditions. Taken together, these findings provide initial,353 yet conclusive support for sleep-mediated modulation of354 tumorigenesis, and suggest that host-dependent immune355 mechanisms constitute a major pathway of such modulatory356 influences.357 The paramount observation of the present study is the358 increased tumor proliferation and marked changes in inva-359 siveness induced by chronic sleep fragmentation exposures360 mimicking several sleep disorders. As such, these observations361 provide initial observations in amurinemodel on the effects of362 perturbed sleep on cancer biology. Furthermore, our study363 offers potential biologic plausibility to some of the findings364 derived from cross-sectional cohorts that have identified epi-365 demiologic associations between perturbed sleep or reduced366 sleep duration to the incidence and overall outcomes of cancer367 (2–17). Thus, by establishing the biologic plausibility of sleep as368 a modulator of the intrinsic properties of cancerous tumors,369 the rationale for further well-controlled epidemiologic and370 intervention studies is warranted. Furthermore, current find-371 ings establish a unique murine model that should allow for372 delineation of the generalizability of sleep fragmentation–373 induced TC1 and 3LLC tumor property modifications to other374 solid and nonsolid clinically relevant tumors, and further375 enable exploration of specific changes in the biologic proper-

377ties of specific cellular componentswithin each of such tumors.378In addition, this experimental paradigm will not only permit379identification of factors involved in tumor proliferation and380invasion, but may also facilitate the investigation of whether381sleep perturbations alter metastatic potential and tumor resis-382tance to therapeutic regimens.383The mechanisms mediating the increased tumor-prolif-384erative rates induced by implementation of sleep fragmen-385tation are unclear and likely diverse. Importantly, sleep386fragmentation–induced effects occurred in both TC-1 and3873LLC tumors, indicating that increased tumor cell-prolifer-388ative rates resulting from perturbations in sleep may not be389specific to a single cancerous cell type, and are potentially390applicable to most solid tumors. Of note, the increased size391of the tumors under sleep fragmentation conditions was392also accompanied by more extensive and prominent areas of393tumor core necrosis, further attesting to the changes in394tumorigenesis elicited by the underlying sleep fragmenta-395tion. In parallel with the accelerated tumor expansion under396sleep fragmentation conditions, increased tumor invasion to397surrounding tissues was observed following both flank and398thigh injections of TC-1 cells, with the latter thigh injections399allowing for more accurate confirmation of the enhanced400invasion process. Here again, multiple mechanisms have401been described about the adaptive strategies that enable402cell invasion across the physical outer boundaries of403tumors. Therefore, an extensive and exhaustive survey of404such mechanisms would be beyond the scope of current405work. Notwithstanding, the unique clustering of TAM in406the periphery of sleep fragmentation–exposed tumors

A SC1

×20 ×20 ×20 ×20

×40 ×40 ×40 ×40

SC2 SF1 SF2

SC1

6 h Postinjection 12 h Postinjection 24 h Postinjectioni ii

SC2

SF1

SC

4

3

2

1

0

* *SF

6 h

Tu

mo

r fl

uo

resce

nc

e (

×1

08)

12 h 24 h

SF2

SF3

B

Figure 2. Sleep fragmentation (SF)induces increased tumorinvasiveness: (A) representativeH&E sections of the TC1 tumorsand the adjacent muscular tissue(cells were injectedsubcutaneously in the thigh area),increased invasiveness wasapparent in SF-C57/B6 tumorscompared with SC-C57/B6conditions, with obviouspenetration of surrounding tumorcapsule and extension/infiltrationinto adjacent muscle. B, i,representative live MMP imaging(n ¼ 6/group) showing statisticallysignificant increases in MMPactivity in SF-C57/B6 tumors vs.CS-C57/B6. (ii) Significantdifferences in MMP activitycontinued to be apparent after 12and 24 hours, with higher activity insleep fragmentation versus sleepconditions (SC) tumors. �, P < 0.05.



409 prompted us to explore whether such changes were poten-410 tially accountable for the more aggressive tumors during411 sleep fragmentation conditions. Indeed, the effects of altera-412 tions in TAM polarity and in innate immunity on tumor413 biologic properties including growth trajectory and inva-414 siveness to adjacent tissues have been extensively explored,415 and our current findings concur with such putative func-416 tions (38–41). As indicated, the number of TAMs was417 markedly increased in sleep fragmentation–exposed mice,418 and the differences were heterotopically distributed, with419 tumors from sleep fragmentation–exposed mice displaying420 preferential M2-type TAM localization in peripheral regions421 of the tumor. In contrast, sleep conditions–derived tumors422 had increased M1-type TAM, and these were located, in423 their greatest proportion within the core of the tumor.424 Therefore, we infer that the changes in macrophage polarity425 and location reflect some of the changes in tumor behavior.426 However, it remains unclear what mechanisms are opera-

428tionally activated in the context of how sleep fragmentation429elicits (i) a shift in TAM polarity within the tumor, or (ii)430whether sleep fragmentation fosters increased migration of431TAM from the circulation into the periphery. We should432point out that the overall changes in TAM polarity were433unexpected, particularly when considering the marked434increases in M1 macrophage markers and in total numbers435of macrophages that occur in visceral adipose tissues436during sleep fragmentation (40). It is further possible that437inflammatory changes in adipose tissues surrounding the438tumors that resemble the changes in visceral white adipose439tissue depots during sleep fragmentation may play a role in440the shift of macrophages from the core to the periphery of441the tumor, and further account for the M1/M2 phenotype442shifts found here. Indeed, recent interest on the contribu-443tion of the adipose tissues surrounding tumors to tumor444biologic processes such as promotion of proliferation and445invasion has recently emerged (43–45). Thus, altered sleep

A

DSCT

AM

s/G

ram

tu

mo

r ti

ss

ue

(×

10

6)

TA

Ms

/wh

ole

tu

mo

r (×

10

6)12

9

6

3

0

12

9

6

3

0

*

*

SF

SCi ii

iii iv

SFSC

×2 ×2 ×2×10 ×10

×2 ×20 ×20×10 ×10

SF

SC

A B C A B C

SF

SC SF

SC SF

B

E F

CFigure 3. TAM counts anddistribution: (A) significantincreases in TAM counts in tumortissue occurred in sleepfragmentation (SF)whencomparedwith sleep conditions (SC) (�3-foldincrease). B, almost 5-fold increasein TAM counts emerged whencomparing the number of TAM/whole tumor in sleep fragmentationversus sleep conditions. C,representative FACS count ofTAMs; CD45þ, CD11bþ, F4/80þ

cells. D, immunofluorescenceimaging of whole tumor sectionsfrom both sleep fragmentation andsleep conditionswith imagesA andC representing both sides of thetumor capsule and B representingthe tumor core; those imagesshowing that TAMs arepreferentially distributed and inclose proximity to the tumorcapsule in sleep fragmentationconditions, whereas in sleepconditions, increased tumor corelocation of TAMs was found (n¼ 6;blue, Hoechst-Nuclei; red, CD45þ;Green, F4/80þ). E, representativetumor section stained withhematoxylin and F4/80þ illustratinga preferential distribution of TAMswith more TAMs in the tumorcapsule in sleep fragmentation–exposed mice when comparedwith sleep conditions (n ¼ 10;brown, F4/80þ cells). F,representative endocan stainedsections showing higherimmunoreactivity in the tumorcapsule in sleep fragmentationconditions when compared withsleep conditions; thiswas apparentin lowmagnification (i and ii), aswellas at higher magnification (iii; iv;n¼ 6; brown, Endocan). �,P < 0.05.Q18

Hakim et al.


448 in the host may trigger a complex time-dependent cascade449 of activation and inactivation of biologically relevant450 pathways both systemically and regionally (46), including

452peritumoral fat and intratumoral cellular substrates that453cumulatively orchestrate changes in tumor growth and local454invasiveness.

A

SC

C57/B6 TLR4-KO

SC-C

57/B

6

SF-C57

/B6

SF-TLR

4-KO

SC-T

LR4-

KO

SC-M

YD88

-KO

SF-MYD

88-K

O

SF-TRIF

-KO

SC-T

RIF

-KO

C57/B6 TLR4-KO

C57/B6 TLR4-KOC57/B6 TLR4-KO

CD

20

6/C

D8

6 r

ati

oT

LR

4 M

FI

Tu

mo

r w

eig

ht

(g)

Tu

mo

r w

eig

ht

(g)

Tu

mo

r vo

lum

e (

mm

3)

TA

Ms

/Tu

mo

r (×

10

6)

Ge

ne

ex

pre

ss

ion

fold

in

cre

as

e

0.25

0.20

0.15

0.10

0.05

0.00

1,500

1,000

500

0

1,500

1,000

500

0

1,500

1,000

500

0

1,500

1,000

500

0

15

10

5

0

5

4

3

2

1

0

*

*

**

*

*

*

*

***

**

*

****

** ** ** **

****

*

*

*

SF

SC

IL-1

2bTG

F-bIL

-10

Arg

1

Fizz1

Mrc

1

INF-g

iNOS

SF

SCSF

SCSF

SCSF

B

C D

E

G

F

Figure 4. TAM polarity and TLR4signaling: (A) TAM (CD45þ CD11bþ

F4/80þ) flow-cytometric assessmentfor the proportion of CD 86high, a M1marker or CD 206high, a M2 marker,showing a shift toward increased M2marker expression in TAMs of SF-C57/B6 tumors (n ¼ 6). B, RT-PCRgene expression analysis of TAMsshowed a shift in polarity in sleepfragmentation (SF) tumorswith higherexpression of M2 markers (Fizz1,Arg1, andMrc1) when comparedwithsleepconditions (SC) (n¼6). C, TAMsfrom sleep fragmentation–exposedmice expressed higher levels of TLR4compared with sleep conditions–exposed mice (n ¼ 6). D and E, SC-TLR4�/� mice injected with TC1tumor cells had a significantlyreduced tumor weight and size whencompared with SC-C57/B6 mice(n ¼ 20/group). Moreover, theaccelerated growth and increasedsize induced by sleep fragmentationexposures were completelyabrogated in SF-TLR4�/� mice(n ¼ 20). F, FACS assessment ofTAMs (CD45þ CD11bþ F4/80þ)counts in tumors from TLR4�/�

compared with C57/B6 conditionsshows complete abrogation of theeffect of sleep fragmentation, withsignificant reductions in TAMcount inSF-TLR4�/� mice (n ¼ 20). G, tumorsweight at enucleation after 28 days ofTC1 tumor cells injection in C57/B6wild-type, TLR4�/�, MYD88�/� andTRIF�/� mice, shows that tumors inSF-MYD88�/� and SF-TRIF�/� micehad reduced tumor weight comparedwith SF-C57/B6, but still significantlylarger than in SF-TLR4�/� mice(n ¼ 12/group). In addition, small,albeit statistically significantdifferences between sleepfragmentation and sleep conditionsfor MYD88 and TRIF null miceremained. �, P < 0.05 whencomparing sleep conditions versussleep fragmentation conditions;��, P < 0.05 when comparing differentstrains.



457 Similarly, the increased TLR4 expression in tumors excised458 from sleep fragmentation–exposed mice could either reflect459 selective migration of M2-TLR4þ TAM cells to these tumor460 regions from the systemic circulation, or changes in macro-461 phage polarity, in macrophage TLR4 expression, or in migra-462 tion of M1 from the core to the tumor periphery during sleep463 fragmentation. Accordingly, genetic ablation of TLR4 in mice464 resulted in major curtailment of not only tumor growth,465 but also of the differences between sleep fragmentation and466 sleep conditions. However, the sleep fragmentation–sleep467 conditions differences in tumor size persisted in either468 MYD88�/� and TRIF�/� mice, suggesting that either both469 TLR4 signaling pathways are required for sleep fragmenta-470 tion–induced effect on tumor growth, or that the differences471 between sleep fragmentation and sleep conditions are not472 distinguishable once tumor growth is so markedly reduced473 by TLR4 genetic ablation. As with all other observations474 reported herein, transfer of the findings from an in vivomurine475 model to an in vitro model is obviously impossible, thereby476 hampering our ability to study mechanisms of sleep fragmen-477 tation effects in greater detail. We should also point out that478 the mechanisms underlying sleep fragmentation–induced479 activation of TLR4 signaling in macrophages, and the changes480 in TAMpolarity are completely unknown, and this specific area481 will have to be explored in the future.482 In summary, the present study conclusively demonstrates483 that perturbed sleep leads to major changes in tumorigenesis,484 characterized by increased tumor cell proliferation and inva-485 sion. Alterations in TAM phenotypes, particularly in the tumor486 periphery, and in TLR4 expression in TAM further suggest that487 sleep fragmentation–induced effects on tumor growth and488 invasion may be mediated by host-related responses, partic-

490ularly those involving innate immunity, and that improved491understanding of such pathways may permit improved ther-492apeutic interventions. Considering the high prevalence of sleep493disorders and cancer in middle age or older populations, there494are far reaching implications to current findings about poten-495tial adverse outcomes in patients in whom the two conditions496coexist.

497Disclosure of Potential Conflicts of Interest Q20

498H. Shirwan is employed as a CSO in ApoVax, Inc., has ownership interest499(including patents) in ApoVax, Inc., and is a consultant/advisory board member500of ApoVax, Inc. No potential conflicts of interest were disclosed by the other501authors.

502Authors' Contributions503Conception and design: F. Hakim, Q21J. Zheng, A. Khlayfa, D. Gozal504Development of methodology: F. Hakim, Y. Wang, S.X.L. Zhang, J. Zheng,505A. Carreras, D. Gozal506Acquisition of data (provided animals, acquired and managed patients,507provided facilities, etc.):F. Hakim, Y.Wang, J. Zheng, A. Carreras, I. Almendros,508D. Gozal509Analysis and interpretation of data (e.g., statistical analysis, biostatistics,510computational analysis): F. Hakim, Y.Wang, S.X.L. Zhang, J. Zheng, A. Khlayfa,511H. Shirwan, I. Almendros, D. Gozal512Writing, review, and/or revision of themanuscript: F. Hakim, Y. Wang, S.X.513L. Zhang, D. Gozal514Administrative, technical, or material support (i.e., reporting or orga-515nizing data, constructing databases): E.S. Yolcu, H. Shirwan, D. Gozal516Study supervision: F. Hakim, Y. Wang, D. Gozal

517Grant Support518D. Gozal is supported by NIH grants HL-065270 and HL-086662.519The costs of publication of this article were defrayed in part by the payment of520page charges. This article must therefore be hereby marked advertisement in521accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

522Received October 22, 2013; revised December 5, 2013; accepted December 26,5232013; published OnlineFirst xx xx, xxxx.

524 References525 1. Abrams B. Cancer and sleep apnea—the hypoxia connection. Med526 Hypotheses 2007;68:232.Q22

527 2. Kripke DF, Garfinkel L, Wingard DL, Klauber MR, Marler MR. Mortality528 associated with sleep duration and insomnia. Arch Gen Psychiatry529 2002;59:131–6.530 3. Verkasalo PK, Lillberg K, Stevens RG, Hublin C, Partinen M, Kosken-531 vuo M, et al. Sleep duration and breast cancer: a prospective cohort532 study. Cancer Res 2005;65:9595–600.533 4. Patel SR, Ayas NT, Malhotra MR, White DP, Schernhammer ES,534 Speizer FE, et al. A prospective study of sleep duration and mortality535 risk in women. Sleep 2004;27:440–4.536 5. Pinheiro SP, Schernhammer ES, Tworoger SS, Michels KB. A pro-537 spective study on habitual duration of sleep and incidence of breast538 cancer in a large cohort of women. Cancer Res 2006;66:5521–5.539 6. Chung SA, Wolf TK, Shapiro CM. Sleep and health consequences540 of shift work in women. J Womens Health (Larchmt) 2009;18:541 965–77.542 7. Gallicchio L, Kalesan B. Sleep duration and mortality: a systematic543 review and meta-analysis. J Sleep Res 2009;18:148–58.544 8. Ikehara S, Iso H, Date C, Kikuchi S, Watanabe Y, Wada Y, et al.545 Association of sleep duration with mortality from cardiovascular dis-546 ease andother causes for Japanesemen andwomen: the JACCstudy.547 Sleep 2009;32:295–301.548 9. Kakizaki M, Kuriyama S, Sone T, Ohmori-Matsuda K, Hozawa A,549 Nakaya N, et al. Sleep duration and the risk of breast cancer: the550 Ohsaki Cohort Study. Br J Cancer 2008;99:1502–5.

55210. KripkeDF, LangerRD, Elliott JA,KlauberMR,RexKM.Mortality related553to actigraphic long and short sleep. Sleep Med 2011;12:28–33.55411. Liang JA, Sun LM, Muo CH, Sung FC, Chang SN, Kao CH. Non-apnea555sleep disorders will increase subsequent liver cancer risk—a nation-556wide population-based cohort study. Sleep Med 2012;13:869–74.55712. McElroy JA, Newcomb PA, Titus-Ernstoff L, Trentham-Dietz A, Hamp-558ton JM, Egan KM. Duration of sleep and breast cancer risk in a large559population-based case-control study. J Sleep Res 2006;15:241–9.56013. Sateia MJ, Lang BJ. Sleep and cancer: recent developments. Curr561Oncol Rep 2008;10:309–18.56214. Thompson CL, Larkin EK, Patel S, Berger NA, Redline S, Li L. Short563duration of sleep increases risk of colorectal adenoma. Cancer5642011;117:841–7.56515. Tu X, Cai H, Gao YT, Wu X, Ji BT, Yang G, et al. Sleep duration and its566correlates in middle-aged and elderly Chinese women: the Shanghai567Women's Health Study. Sleep Med 2012;13:1138–45.56816. Jiao L, Duan Z, Sangi-Haghpeykar H, Hale L, White DL, El-Serag HB.569Sleep duration and incidence of colorectal cancer in postmenopausal570women. Br J Cancer 2013;108:213–21.57117. Zhang X, Giovannucci EL, Wu K, Gao X, Hu F, Ogino S, et al. Associa-572tions of self-reported sleepduration and snoringwith colorectal cancer573risk in men and women. Sleep 2013;36:681–8.57418. Savvidis C, Koutsilieris M. Circadian rhythm disruption in cancer575biology. Mol Med 2012;18:1249–60.57619. Sigurdardottir LG, Valdimarsdottir UA, Fall K, Rider JR, Lockley SW,577Schernhammer E, et al. Circadian disruption, sleep loss, and prostate

Hakim et al.


580 cancer risk: a systematic review of epidemiologic studies. Cancer581 Epidemiol Biomarkers Prev 2012;21:1002–11.582 20. Campos-Rodriguez F, Martinez-Garcia MA, Martinez M, Duran-Can-583 tolla J, PenaMdeL,MasdeuMJ, et al. Associationbetweenobstructive584 sleep apnea and cancer incidence in a large multicenter Spanish585 cohort. Am J Respir Criti Care Med 2013;187:99–105.586 21. Nieto FJ, Peppard PE, Young T, Finn L, Hla KM, Farre R. Sleep-587 disordered breathing and cancer mortality: results from theWisconsin588 Sleep Cohort Study. A J Respir Crit Care Med 2012;186:190–4.589 22. Almendros I, Montserrat JM, Ramirez J, Torres M, Duran-Cantolla J,590 NavajasD, et al. Intermittent hypoxia enhances cancer progression in a591 mouse model of sleep apnoea. Eur Respir J 2012;39:215–7.592 23. Almendros I, Montserrat JM, Torres M, Dalmases M, Cabanas ML,593 Campos-Rodriguez F, et al. Intermittent hypoxia increases melanoma594 metastasis to the lung in amousemodel of sleep apnea. Respir Physiol595 Neurobiol 2013;186:303–7.596 24. Bhaskara VK, Mohanam I, Rao JS, Mohanam S. Intermittent hypoxia597 regulates stem-like characteristics and differentiation of neuroblasto-598 ma cells. PLoS ONE 2012;7:e30905.599 25. Karoor V, LeM,Merrick D, Fagan KA, Dempsey EC,Miller YE. Alveolar600 hypoxia promotes murine lung tumor growth through a VEGFR-2/601 EGFR-dependent mechanism. Cancer Prev Res 2012;5:1061–71.602 26. Liu Y, Song X, Wang X, Wei L, Liu X, Yuan S, et al. Effect of chronic603 intermittent hypoxia on biological behavior and hypoxia-associated604 gene expression in lung cancer cells. J Cell Biochem 2010;111:605 554–63.606 27. Malec V, Gottschald OR, Li S, Rose F, SeegerW, Hanze J. HIF-1 alpha607 signaling is augmented during intermittent hypoxia by induction of the608 Nrf2 pathway in NOX1-expressing adenocarcinoma A549 cells. Free609 Radic Biol Med 2010;48:1626–35.610 28. Diers AR, Vayalil PK, Oliva CR, Griguer CE, Darley-Usmar V, Hurst DR,611 et al. Mitochondrial bioenergetics of metastatic breast cancer cells in612 response to dynamic changes in oxygen tension: effects of HIF-613 1alpha. PLoS ONE 2013;8:e68348.614 29. Ramesh V, Nair D, Zhang SX, Hakim F, Kaushal N, Kayali F, et al.615 Disrupted sleep without sleep curtailment induces sleepiness and616 cognitive dysfunction via the tumor necrosis factor-alpha pathway.617 J Neuroinflamm 2012;9:91.618 30. Zhang SX, Khalyfa A, Wang Y, Carreras A, Hakim F, Neel BA, et al.619 Sleep fragmentation promotes NADPH oxidase 2-mediated adipose620 tissue inflammation leading to insulin resistance in mice. Int J Obes.621 2013 Jul 30. [Epub ahead of print].Q23

622 31. Fukuda K, Kobayashi A, Watabe K. The role of tumor-associated623 macrophage in tumor progression. Front Biosci (Schol Ed) 2012;4:624 787–98.625 32. Ruffell B, Affara NI, Coussens LM. Differential macrophage program-626 ming in the tumor microenvironment. Trends Immunol 2012;33:119–26.

62833. Kaushal N, Ramesh V, Gozal D. Human apolipoprotein E4 targeted629replacement in mice reveals increased susceptibility to sleep disrup-630tion and intermittent hypoxia. Am J Physiol Regul, Integrat Comp631Physiol 2012;303:R19–29.63234. Gialeli C, Theocharis AD, Karamanos NK. Roles of matrix metallopro-633teinases in cancer progression and their pharmacological targeting.634FEBS J 2011;278:16–27.63535. KessenbrockK,PlaksV,WerbZ.Matrixmetalloproteinases: regulators636of the tumor microenvironment. Cell 2010;141:52–67.63736. Hua D, Liu MY, Cheng ZD, Qin XJ, Zhang HM, Chen Y, et al. Small638interfering RNA-directed targeting of Toll-like receptor 4 inhibits639human prostate cancer cell invasion, survival, and tumorigenicity. Mol640Immunol 2009;46:2876–84.64137. Yang H, Zhou H, Feng P, Zhou X, Wen H, Xie X, et al. Reduced642expression of Toll-like receptor 4 inhibits human breast cancer cells643proliferation and inflammatory cytokines secretion. J Exp Clin Cancer644Res 2010;29:92.64538. Gajewski TF, Schreiber H, Fu YX. Innate and adaptive immune cells in646the tumor microenvironment. Nat Immunol 2013;14:1014–22.64739. Green CE, Liu T, Montel V, Hsiao G, Lester RD, Subramaniam S, et al.648Chemoattractant signaling between tumor cells and macrophages649regulates cancer cell migration, metastasis and neovascularization.650PLoS ONE 2009;4:e6713.65140. Obeid E, Nanda R, Fu YX, Olopade OI. The role of tumor-associated652macrophages in breast cancer progression (review). Int J Oncol6532013;43:5–12.65441. Takahashi T, Ibata M, Yu Z, Shikama Y, Endo Y, Miyauchi Y, et al.655Rejection of intradermally injected syngeneic tumor cells frommice by656specific elimination of tumor-associatedmacrophages with liposome-657encapsulated dichloromethylene diphosphonate, followed by induc-658tion of CD11b(þ)/CCR3(-)/Gr-1(-) cells cytotoxic against the tumor659cells. Cancer Immunol, Immunother 2009;58:2011–23.66042. Wang Y, Carreras A, Lee S, Hakim F, Zhang SX, Nair D, et al. Chronic661sleep fragmentation promotes obesity in young adult mice. Obesity.6622013 Sep 13. [Epub ahead of print].66343. Howe LR, Subbaramaiah K, Hudis CA, Dannenberg AJ. Molecular664pathways: adipose inflammation as a mediator of obesity-associated665cancer. Clin Cancer Res 2013;19:6074–83.66644. Nieman KM, Romero IL, Van Houten B, Lengyel E. Adipose tissue and667adipocytes support tumorigenesis and metastasis. Biochim Biophys668Acta 2013;1831:1533–41.66945. Wagner M, Dudley AC. A three-party alliance in solid tumors: adipo-670cytes, macrophages and vascular endothelial cells. Adipocyte6712013;2:67–73.67246. Gharib SA, Khalyfa A, Abdelkarim A, Bhushan B, Gozal D. Integrative673miRNA-mRNA profiling of adipose tissue unravels transcriptional674circuits induced by sleep fragmentation. PLoS ONE 2012;7:e37669.



AUTHOR QUERIES

AUTHOR PLEASE ANSWER ALL QUERIES

Q1: Page: 1: AU: Per journal style, genes, alleles, loci, and oncogenes are italicized;proteins are roman. Please check throughout to see that thewords are styled correctly.AACR journals have developed explicit instructions about reporting results fromexperiments involving the use of animal models as well as the use of approved geneand protein nomenclature at their first mention in the manuscript. Please review theinstructions at http://www.aacrjournals.org/site/InstrAuthors/ifora.xhtml#gene-nomen to ensure that your article is in compliance. If your article is not in compliance,please make the appropriate changes in your proof.

Q2: Page: 1: Author: Please verify the drug names and their dosages used in the article.

Q3: Page: 1: Author: Please verify the affiliations and their corresponding author links.

Q4: Page: 1: Author: Please verify the corresponding author details.

Q5: Page: 1: Author: Note that as per AACR, author created two-letter abbreviations aregenerally not allowed (except the ones mentioned in the manual). Therefore, "SF" and"SC" have been expanded throughout the article (apart from constructs). Please verify.

Q6: Page: 2: Author: Units of measurement have been changed here and elsewhere in thetext from"M" to "mol/L", and relatedunits, suchas "mmol/L" andmmol/L," infigures,legends, and tables in accordance with journal style, derived from the Council ofScience Editors Manual for Authors, Editors, and Publishers and the Syst�eme inter-national d’unit�es. Please note if these changes are not acceptable or appropriate in thisinstance.

Q7: Page: 2: Author: Please verify the changesmade in the sentence "Twelve C57/B6male. . . disease-related MMPs" for correctness.

Q8: Page: 2: Author: Please define "PFA."

Q9: Page: 2: Author: Please verify the change of "mM" to "mm" in the sentence "Cells werefiltered . . . manufacturer’s procedure" for correctness.

Q10: Page: 2: Author: Please verify the expansion of "OCT" for correctness.

Q11: Page: 2: Author: Please verify the changes made in the sentence "Of note, 10 mm . . .with F4/80" for correctness.

Q12: Page: 3: Author: Please verify the changes made in the sentence "Sections were then. . . for each wash" for correctness.

Q13: Page: 3: Author: Please verify the presentation of "BioLegend" for correctness.

Q14: Page: 3: Author: Please verify the changes made in the sentence "Of note, 500 . . . forRT-PCR" for correctness.

Q15: Page: 3: Author: Please verify the changes made in the sentence "These Ct values . . .was calculated" for correctness.

Q16: Page: 3: Author: Please verify the changes made in the sentence "The followingTaqMan . . . Mm02529782_s1" for correctness.

Q17: Page: 4: Author: Author: Please confirm quality/labeling of all images includedwithin this article. Thank you.

Q18: Page: 6: Author: Please provide an in-text citation of Fig. 3C.

Q19: Page: 4: Author: Please verify the changes made in the sentence "To further under-stand . . . a M2 marker" for correctness.

Q20: Page: 8:Author:AU/PE: The conflict-of-interest disclosure statement that appears inthe proof incorporates the information from forms completed and signed off on byeach individual author. No factual changes can be made to disclosure information atthe proof stage. However, typographical errors or misspelling of author namesshould be noted on the proof and will be corrected before publication. Please noteif any such errors need to be corrected. Is the disclosure statement correct?

Q21: Page: 8: Author: The contribution(s) of each author are listed in the proof under theheading "Authors’ Contributions." These contributions are derived from formscompleted and signed off on by each individual author. As the corresponding author,you are permitted to make changes to your own contributions. However, because allauthors submit their contributions individually, you are not permitted to makechanges in the contributions listed for any other authors. If you feel strongly thatan error is beingmade, then youmay ask the author or authors in question to contactus about making the changes. Please note, however, that the manuscript would beheld from further processing until this issue is resolved.

Q22: Page: 8: Author: Note that Refs. 36, 37, and 42 are given in the reference list but notcited in the text. Please provide an appropriate citation for these references in the text.

Q23: Page: 9: Author: Note that Refs. 30, 42, and 43 have been updated as per PubMed.Please verify.

AU: Below is a summary of the name segmentation for the authors according to our records.The First Name and the Surname data will be provided to PubMedwhen the article is indexedfor searching. Please check eachname carefully and verify that theFirstName andSurname arecorrect. If a name is not segmented correctly, please write the correct First Name and Surnameon this page and return it with your proofs. If no changes are made to this list, we will assumethat the names are segmented correctly, and the names will be indexed as is by PubMed andother indexing services.

First Name Surname

Fahed Hakim

Yang Wang

Shelley X.L. Zhang

Jiamao Zheng

Esma S. Yolcu

Alba Carreras

Abdelnaby Khlayfa

Haval Shirwan

Isaac Almendros

David Gozal

fragmented sleep accelerates tumor growth and progression ...€¦ · 1 q1 integrated systems and...

Documents